MALEBRANCHE
ARGUMENTS OF THE PHILOSOPHERS
The purpose of this series is to provide a contemporary assessment and his...
77 downloads
2032 Views
1MB Size
Report
This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form
MALEBRANCHE
ARGUMENTS OF THE PHILOSOPHERS
The purpose of this series is to provide a contemporary assessment and history of the entire course of philosophical thought. Each book contains a detailed, critical introduction to the work of a philosopher or school of major influence and significance. Also available in the series: AQUINAS Eleonore Stump
* NIETZSCHE Richard Schacht
* DESCARTES Margaret D. Wilson
PLATO Justin Gosling
* HEGEL M.J. Inwood
* PLOTINUS Lloyd P. Gerson
* HUME Barry Stroud
* ROUSSEAU Timothy O’Hagan
KANT Ralph C.S. Walker
* THE PRESOCRATIC PHILOSOPHERS Jonathan Barnes
KIERKEGAARD Alastair Hannay * LOCKE Michael Ayers * KARL MARX Allen Wood * MERLEAU-PONTY Stephen Priest
* SANTAYANA Timothy L.S. Sprigge * THE SCEPTICS R.J. Hankinson * WITTGENSTEIN, 2nd edition Robert Fogelin * also available in paperback
MALEBRANCHE
Andrew Pyle
First published 2003 by Routledge 11 New Fetter Lane, London EC4P 4EE Simultaneously published in the USA and Canada by Routledge 29 West 35th Street, New York, NY 10001 Routledge is an imprint of the Taylor & Francis Group This edition published in the Taylor & Francis e-Library, 2003. © 2003 Andrew Pyle All rights reserved. No part of this book may be reprinted or reproduced or utilised in any form or by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying and recording, or in any information storage or retrieval system, without permission in writing from the publishers. British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library Library of Congress Cataloging in Publication Data Pyle, Andrew. Malebranche / Andrew Pyle. p. cm. – (The arguments of the philosophers) Includes bibliographical references and index. 1. Malebranche, Nicolas, 1638–1715. I. Title. II. Series. B1897 .P95 2002 194–dc21 2002032459 ISBN 0-203-41770-4 Master e-book ISBN
ISBN 0-203-41915-4 (Adobe eReader Format) ISBN 0–415–28911–4 (Print Edition)
In memory of David Hirschmann (1939–97), teacher, colleague and friend, lover of the good things of life – French civilisation, slow food, red wine, bridge, and the rationalist philosophy of the seventeenth century. My one regret regarding this book is that he will never read it, although it owes so much to him in so many ways.
CONTENTS
CONTENTS
x xii
Abbreviations and editions Preface 1
2
3
4
Introduction Life and works 1 Descartes and Augustine Faith and reason 9 Difficulties 13
1 5
Tensions in Cartesian metaphysics Richard Watson’s Downfall of Cartesianism Descartes on ideas 21 Later Cartesian accounts of ideas 28 Descartes on causation 33 Later Cartesian accounts of causation 39
18 18
The Vision in God The argument for ideas 47 The eliminative argument for the Vision in God The argument from properties 57 Intelligible extension 61 Efficacious ideas 67 Nadler’s Malebranche 71
47 50
The dispute with Arnauld over the nature of ideas Arnauld’s Vraies et Fausses Idées 75 Malebranche’s Réponse 79 What is at stake? Arnauld’s version of the debate 83 What is at stake? Malebranche’s version of the debate 85 Deeper reasons 89 vii
74
CONTENTS
5
Occasionalism and continuous creation How not to think of occasionalism 96 The idea of necessary connection 98 Particular causal relations 101 Continuous creation 111 Continuous creation and volontés générales Is this the best of all possible worlds? 118 Objections and replies 121
96
114
6
Malebranche’s modifications of Cartesian physics Malebranche’s Cartesian inheritance 131 Malebranche’s philosophical physics 141 Malebranche’s first revision of Descartes: drop the force of rest 144 Malebranche’s second revision of Descartes: abandon the scalar conservation principle 147 Malebranche’s third revision of Descartes: reject hard bodies 150 Is Malebranche an empiricist malgré lui? 154
131
7
Malebranche’s biology Nature and supernature 158 Descartes’ naturalistic cosmogony 160 Worlds and organisms in Malebranche 161 The theory of pre-existence 166 Creation and special providence 173 The problem of monsters 176 The maternal imagination and original sin 180 Conclusion: reason and experience in Malebranche’s biology 182
158
8
Malebranche on the soul and self-knowledge Descartes’ arguments for the immateriality of the soul We lack a clear idea of the soul 188 Divine and human knowledge 191 The via negativa route to dualism 195 Personal immortality under threat 198 Metaphysics and introspection 200 The sciences of psychology 203 Two final questions 207
viii
186 186
CONTENTS
9
Malebranche on freedom, grace and the will Philosophical background: continuous creation and the Vision in God 209 Theological background: Augustine against the Pelagians Freedom and grace in the Recherche 215 The Traité de la Nature et de la Grâce 221 The controversy with Arnauld 227 The prémotion physique 230
10 The downfall of Malebranchism Epistemological difficulties 235 Metaphysical difficulties 242 Ethical difficulties 246 Theological difficulties 252 Malebranche’s influence 257
209
211
234
262 279 286
Notes Bibliography Indexes
ix
ABBREVIATIONS AND EDITIONS
A B B R E V I AT I O N S AND EDITIONS
Works by Malebranche I have worked entirely with the Oeuvres Complètes de Malebranche, ed. André Robinet, 20 Volumes, Paris, Vrin, 1958–78. References are given to OCM, by volume and page. Individual works by Malebranche are referred to by means of the following abbreviations. R.V.
Recherche de la Vérité, ou l’on traite de la nature de l’esprit de l’homme et de l’usage qu’il en doit faire pour éviter l’erreur dans les sciences (1674). OCM Volumes 1 and 2. Ecl. Éclaircissements sur la Recherche de la Vérité (1678). OCM Vol. 3. C.C. Conversations Chrétiennes (1677). OCM Vol. 4. T.N.G. Traité de la Nature et de la Grâce (1680). OCM Vol. 5. C.A. Correspondance avec M. Arnauld (1684–87). OCM Vols 6– 9. M.C. Méditations Chrétiennes et Métaphysiques (1683). OCM Vol. 10. T.M. Traité de Morale (1684). OCM Vol. 11. E.M. Entretiens sur la Métaphysique et sur la Religion (1688). OCM Vol. 12. E. Mort. Entretiens sur la Mort (1696). OCM Vol. 13. T.A.D. Traité de l’Amour de Dieu (1698). OCM Vol. 14. Lamy. Trois Lettres au Père Lamy (1698). OCM Vol. 14. Chin. Entretien d’un Philosophe Chrétien et d’un Philosophe Chinois (1707). OCM Vol. 15. Pre.Ph. Réflexions sur la Prémotion Physique (1715). OCM Vol. 16. Rep Régis. Réponse a M. Régis (1693). OCM Vol. 17–1. Def. la Ville Défense de l’Auteur de la Recherche de la Vérité contre l’Accusation de Mons. de la Ville (1682). OCM Vol. 17–1.
x
A B B R E V I AT I O N S A N D E D I T I O N S
Math. Corr.
Mathematica. OCM Vol. 17–2. Correspondance et Actes. OCM Vols 18 and 19.
English translations For the Search after Truth and the Dialogues on Metaphysics and Religion I have made use of the existing English translations, both of which have recently appeared in the Cambridge Texts in the History of Philosophy. Thanks are due to Professor Lennon for permission to use his English translation of the Search, and to Cambridge University Press and Professors Jolley and Scott for permission to use their English translation of the Dialogues. For the Treatise on Nature and Grace I have sometimes been able to use the recent translation by Patrick Riley. Since his translation is of the short first edition of 1680, I have been obliged to use the Oeuvres Complètes for material added to later editions. Nicolas Malebranche, The Search After Truth, translated and edited by Thomas Lennon and Paul Olscamp. Plus Elucidations of the Search after Truth, translated and edited by Thomas Lennon, Cambridge University Press, 1997. Nicolas Malebranche, Dialogues on Metaphysics and on Religion, translated by David Scott and edited by Nicholas Jolley, Cambridge University Press, 1997. Nicolas Malebranche, Treatise on Nature and Grace, translated with an introduction and notes by Patrick Riley, Oxford, Clarendon, 1992. These translations will be referred to as LO, JS, and R respectively. References to the Recherche, the Entretiens and, where appropriate, the Traité will be to both the OCM version and the English translation (e.g. OCM III 163, LO 633). Translations of all other works by Malebranche are my own.
Works of Descartes AT = Oeuvres de Descartes, Charles Adam and Paul Tannery, eds, 12 Volumes (Paris, Cerf, 1897–1910). CSM(K) = The Philosophical Writings of Descartes, translated and edited by John Cottingham, Robert Stoothoff, Dugald Murdoch, and (for Vol. 3) Anthony Kenny, 3 Volumes (Cambridge University Press, 1985–91). References to the works of Descartes will be given both to the original French or Latin, and to the English translation where available (e.g. AT IX 37, CSM II 31).
xi
P R E FA C E
P R E FA C E
This book has been more than a decade in the writing. The idea for it came to me during my two years teaching at the University of Ghana at Legon, from 1986 to 1988. The university’s Balme Library was not well stocked with modern books and journals but contained, by way of compensation, many treasures from the past. In an idle moment during the long vacation I picked up Malebranche’s Recherche de la Vérité, partly to improve my French, partly for its intrinsic interest. I have been wrestling with Malebranche’s philosophy ever since. It is not, I hasten to add, that I share many of his opinions; my own philosophical views could scarcely be further removed from his. But I have no doubt at all that he is a major philosopher, frequently misrepresented and – at least until recently – quite unjustly neglected in the English-speaking world. In the past few years, largely through the efforts of Steven Nadler, Malebranche has enjoyed a small but significant revival among anglophone historians of philosophy. Since my return to Bristol I have finished my study of the twenty volumes of Robinet’s great Oeuvres Complètes de Malebranche, together with a fair selection of the secondary literature in French and English. I have also written articles on various aspects of Malebranche’s philosophy for a number of different occasions and audiences, and these articles have been pressed into service as draft chapters for this book. Apologies are owed to readers for the occasional small repetition that may have escaped the revision process. Draft versions of chapters three and eight were presented to meetings of the British Society for the History of Philosophy, and drafts of chapters four and nine to Tom Sorell’s study group for Seventeenth Century Philosophy. I am grateful to Tom Sorell, John Cottingham, Stuart Brown, Pauline Phemister, Susan James, Sarah Hutton, John Rogers, Ian Tipton, Peter Alexander, Martin Stone and others both for their encouragement and for their objections. Draft versions of
xii
P R E FA C E
chapters six and seven were presented to meetings of the Australian History and Philosophy of Science community at the universities of Sydney and Melbourne in June and July of 1997. I am grateful to Alan Chalmers, Stephen Gaukroger and Keith Hutchison, among others, for helpful questions and criticisms. A major debt of thanks is also owed to my department here in Bristol. A succession of heads of department – Michael Welbourne, David Hirschmann, Chris Bertram and Keith Graham – have provided support and encouragement combined with only the gentlest of pressure. Thanks too to the Research Committee for the period of study leave that has enabled me to complete this book. After seeing off our TQA inspection in March 2001 it was a pleasure to return to the library and the purer joys of research, and to be able to work more or less uninterrupted until October. Most of my colleagues, past and present, have discussed draft chapters at our departmental staff seminars. The influence of these discussions on the final book is too diffuse to document in detail, but no less real for that. So although this project may have been conceived under the tropical sun of Africa, it grew to maturity in the milder climate of the West Country. Thanks are also due to the two anonymous referees chosen by Routledge, both for their praise and for their criticisms. Their praise reassured me that the work would have some value; their criticisms helped me to clarify some key points and eliminate a handful of small but significant errors. These readers will find that their detailed criticisms have been met by a mixed strategy of concession, clarification, and rebuttal. Suggestions for more fundamental re-structuring of the work have been quietly resisted. Once a book has taken its final shape in the mind of its author, it is much easier to address objections to points of detail than to rethink the form of the whole.
xiii
P R E FA C E
xiv
INTRODUCTION
1 INTRODUCTION
Life and works Nicolas Malebranche was born and died in the same year as Louis XIV (1638–1715), and thus belongs firmly to the siècle des lumières celebrated in Voltaire’s history.1 Born in Paris, he was the youngest of the many children of Nicolas Malebranche senior, a counsellor of the king and a senior official in the department of taxation.2 Nicolas junior was a feeble child, and was not expected to live long; as a man, he remained frail and sickly, as if made for the secluded life of an ascetic and an intellectual. The obvious profession for a bright but feeble-bodied child was the Church. From 1654 to 1656 Malebranche studied philosophy at the Collège de la Marché in Paris, then in 1656 he started his theology at the Sorbonne (OCM XVIII 13). According to his first biographer, André, the theology he was taught failed to impress him on two counts. In philosophy, where reason alone should be sovereign, he was fobbed off with the mere authority of Aristotle; in theology, where authority alone should rule, he was offered human (and usually bad) reasoning (OCM XVIII 16). Although originally destined for the secular priesthood, the young Malebranche withdrew into the congregation of the Oratory in 1660, giving up his share of the family property. The Oratorian Order had been founded by Pierre Bérulle in 1611. Its members had no official duties, but were expected to devote their lives to prayer, fasting and works of charity. Dogmatic and mystical theology was permitted, but independent philosophical thinking was associated with the sin of pride, and was officially discouraged: Malebranche’s writings would earn him the bitter hostility of his superiors. As a student, he plodded diligently through vast tomes of Church history, and even began the study of Hebrew, before he found his true vocation. In 1664, the year of his ordination, he encountered Descartes’ Traité de l’Homme (OCM XVIII
1
INTRODUCTION
16). For the young Malebranche, as for so many others, Descartes’ ideas had the force of a revelation. Here, he felt, was the true philosophy, firmly based on clear and distinct ideas, and providing demonstrative arguments with all the rigour of mathematical proofs. For the next few years, he devoured the works of Descartes and his disciples, developing a distinctive philosophical position in which a Cartesian, i.e. mechanistic, natural philosophy is incorporated into a spiritualistic metaphysics derived largely from Augustine.3 As André puts it, ‘The sublime metaphysics of Saint Augustine seemed entirely made for the physics of Mr Descartes, and the physics of Mr Descartes for the metaphysics of Saint Augustine’ (OCM XVIII 51). This modified Cartesianism was to be put into the service of Christian theology. As Gueroult points out, a pagan or a Turk could be a Cartesian: there is nothing specifically Christian about Descartes’ metaphysics and epistemology.4 Malebranche deliberately set out, by contrast, to create a Christian philosophy. His explicitly stated aim was to do for Descartes what Augustine had done for Plato, or Aquinas for Aristotle, i.e. to use the philosophy to provide new explanations for the unchanging dogmas of the faith. This is not merely, of course, a concession to the philosophical fashions of the age. Malebranche thought that, with certain modifications which we shall note, Cartesianism was the true philosophy, so the true philosophy supports – as of course it should – the true religion.5 The fruits of Malebranche’s assimilation of Descartes appeared in the Recherche de la Vérité, published anonymously in 1674.6 The Recherche is a long and rambling work, ostensibly devoted to the causes of human error, although frequently digressing into substantive matters of mathematics, physics, metaphysics, ethics and theology. It deals at length with our various faculties – sense, imagination, the passions, and reason – and with the various ways in which they can lead us astray. Drawing heavily on Descartes’ Fourth Meditation, Malebranche announces two rules for the guidance of the mind that desires to free itself from such errors and arrive at the truth. Once our minds have been properly purged of vulgar errors, Malebranche claims, two great truths will become manifest, opening clear paths into the subjects of metaphysics and epistemology. In metaphysics, we will see clearly that only God can be a true cause, and that what we commonly call ‘natural’ or ‘second’ causes are mere occasions for God to act in accordance with His own self-imposed rules. In epistemology, we will find that human knowledge is only possible by means of the ‘Vision in God’, i.e. if ideas are archetypes in the Creator’s mind, not merely modifications or ‘modes’ of our own souls. These twin theses of occasionalism and the Vision in God constitute 2
INTRODUCTION
the heart of Malebranche’s philosophy. In his later works he would refine these doctrines, provide further explanations of them, and defend them against objections; he would never abandon them.7 The Recherche was, from the first, the subject of heated controversy. The sceptic Simon Foucher took issue with Malebranche’s epistemology, and objected strongly to the mixture of religion and philosophy that is such a striking feature of the book.8 The Jesuits, still wedded to scholastic Aristotelianism, sought to portray Malebranche’s Cartesian exposition of Christian doctrine as dangerously heretical.9 Initially at least, the Jansenists’ response was favourable: Arnauld spoke warmly of the Recherche, and clearly hoped that its author could be drawn into a Cartesian/Augustinian/Jansenist camp, in opposition to the Aristotelian/ Thomist/Jesuit one. The publication in 1678 of the Éclaircissements to the Recherche put an end to this hope. In Éclaircissement XV (OCM III 220–1, LO 666–7), Malebranche argued that an all-perfect being must, by its very nature, act by means of general laws. Since the proof of this proposition is a priori, it is perfectly general in its application. In the Traité de la Nature et de la Grâce (1680), Malebranche’s most controversial work, the application of this thesis to the theological domain of grace is spelt out. The distribution of grace, we are told, is effected by a general law by means of which the (human) soul of Jesus Christ is established as an occasional cause. God, says Malebranche, wills that all men be saved. But He distributes grace to sinners in accordance with the above rule. This means that many men do not receive enough grace to accomplish their salvation, and so are damned. Why does God act in such a way? Because He owes it to Himself to act in a manner that is worthy of His attributes – in this case, His wisdom. As in the realm of nature, so in that of grace, God acts as a general cause rather than a particular cause. The Traité was, in effect, a declaration of war against the Jansenists. At the heart of their theology was the doctrine of ‘efficacious grace’, i.e. the belief that God’s gift of grace always effects the conversion of the soul of the sinner and thus leads to salvation. This requires that God distribute grace by particular volitions, never giving any of the chosen elect less than he or she needs. On Malebranche’s theory, such unfortunate events are common: just as some fields receive too little rain for the crops to grow, so some souls receive too little grace to effect conversion and reconciliation with God. From Arnauld’s point of view, Malebranche tacitly denies God’s omnipotence: he says that God wills that all men be saved, but that this end is not achieved. From Malebranche’s point of view, Arnauld effectively denies God’s universal 3
INTRODUCTION
benevolence, and falls into a position akin to Calvinism. If God acts by particular volitions, and some men are damned, it is because God didn’t give them enough grace, so could not have willed their salvation. After the publication of the Traité, Arnauld published a series of attacks on the philosophy of Malebranche; his attacks and Malebranche’s replies occupy no less than four volumes of the Oeuvres Complètes (OCM VI–IX). Arnauld attacked Malebranche’s philosophy on two separate fronts. In his Vraies et Fausses Idées10 he rejects the Vision in God and defends what he takes to be the orthodox Cartesian view that our ideas are modes (acts) of our own souls, not archetypes in the divine intellect. But why does Arnauld choose to attack Malebranche’s theology by way of his epistemology? The reason is clear.11 The doctrine of the Vision in God allows the human mind to participate, at least to some extent, in the divine intellect, and thus to grasp, at least in some cases, God’s reasons for acting.12 This rationalisation of theology is anathema to Arnauld. For him, whatever God does is right simply because God does it – there are no standards of right and wrong independent of the divine will. In essence, we have here another episode in the endless struggle between the rival voluntarist and intellectualist traditions in Christian theology.13 Arnauld sees Malebranche’s intellectualism as lying behind and supporting his theology, and so launches his first salvo directly against the Vision in God. He goes on, of course, in his subsequent series of letters, to attack the theology of the Traité directly, launching a vigorous polemical exchange with Malebranche that lasted until Arnauld’s death in 1694. Malebranche’s other major philosophical work, the Entretiens sur la Métaphysique, was published in 1688. These dialogues are, without question, Malebranche’s mature philosophical masterpiece. The Recherche is a young man’s book – ambitious in its scope, rambling in its structure, full of digressions, and not always clear in the presentation of its central ideas. It was also subject to substantial revisions, resulting in a stratified text in which inconsistent opinions from different periods are sometimes visible in startling juxtaposition.14 The Entretiens are, by contrast, a model of lucid exposition, with Malebranche’s spokesman Theodore guiding his young pupil Ariste (and thus also the reader) through his central doctrines, addressing difficulties and objections as they arise. If a modern reader has time to read only one work of Malebranche, it should be the Entretiens rather than the Recherche. These works, the Recherche de la Vérité and its Éclaircissements, the Traité de la Nature et de la Grâce, the Correspondance avec Arnauld and the Entretiens sur la Métaphysique, constitute the philosophical core 4
INTRODUCTION
of Malebranche’s work, and will provide the bulk of the material for this study. On occasions we may dip into our author’s lesser works, such as the Conversations Chrétiennes, the Méditations Chrétiennes, the Traité de l’Amour de Dieu, the Prémotion Physique, and the Traité de Morale, but we shall do so only where they cast light on the central texts.
Descartes and Augustine The most cursory and superficial reading of Malebranche’s works will suffice to discover the main influences shaping his philosophy. As he himself explains in a letter to Arnauld, there are only two earlier philosophers who exerted a major formative influence on his thought: Augustine and Descartes (OCM VIII 998). He never acknowledges any debt to the scholastics, although this may not be entirely fair on his part.15 It is important not to think of Augustine and Descartes as the spokesmen, respectively, of faith and reason. For Malebranche, both are voices of reason; where they part company is in their very different conceptions of reason itself.16 Descartes thinks of reason as a natural (albeit doubtless God-given) power of the human soul; Augustine stresses the essential dependence of our minds on the ‘light’ of divine illumination. On this point Malebranche sides with Augustine: all our knowledge is some form or other of revelation. The assimilation of Cartesian and Augustinian philosophy was quite common in the seventeenth century, usually in opposition to the Aristotelianism of the scholastics. Scholastic Aristotelianism is denounced by Malebranche as both bad philosophy and bad theology. It is bad philosophy because it possesses no clear and distinct ideas, and thus falls into vulgar errors regarding causality and the relation between mind and body. It is bad theology because the admission of natural causes risks a slide into pagan naturalism, and the failure to distinguish clearly between mind and body threatens the Christian doctrine of the immortality of the soul. The precarious reconciliation of Aristotelianism and Christianity effected by Thomas Aquinas was never very robust: the tensions which had been present in Thomism from the beginning threatened, by the seventeenth century, to tear scholasticism apart. The time had come, according to Malebranche, to replace scholastic Aristotelianism with a new Christian philosophy derived from Descartes and Augustine. Malebranche adopts from Descartes, first and foremost, the famous method – this is clear from the two opening chapters of Book One of the Recherche (OCM I 39–57, LO 1–11). In Malebranche’s version of the
5
INTRODUCTION
method of doubt, however, the radical scepticism of Descartes’ First Meditation disappears without trace. Malebranche never entertains the slightest doubt of the veridical nature of his clear and distinct ideas. If I have a clear and distinct idea of X that represents it as having feature F, I know that X is F, and know that I know that X is F, with no room for further doubt. Doubt remains appropriate until clear and distinct ideas are grasped; after that, it simply evaporates. Errors arise, for Malebranche as for Descartes, by precipitate acts of judgement, assenting to or dissenting from propositions before we are in a proper position to do so.17 The account of error contained in the Recherche is a sort of extended commentary on Descartes’ Fourth Meditation: error is always our own fault, since it results from an abuse of our liberty. Where matters are obscure, we should suspend judgement and attend more closely, seeking further illumination. Attention itself is, for Malebranche, a sort of prayer, and such prayers are always answered, provided we are sincere and persistent, and the matter itself is one that lies, in principle, within the compass of our understandings. Malebranche also endorses Descartes’ strict metaphysical dualism. Indeed, Malebranche was a more rigorous and thoroughgoing dualist than Descartes himself. In his correspondence with the Princess Elizabeth,18 Descartes invites readers to take the notion of a mind–body union seriously, and suggests that a certain range of psychological powers (sensation, imagination, memory) are powers of the mind–body union rather than of the separate incorporeal soul.19 Malebranche will have none of this. In Book Two of the Recherche he tells us clearly that the notion of a substantial union of mind and body is unintelligible. All that the notion of ‘union’ can mean here is that God has established regular and lawlike correlations between states of the brain, on the one hand, and states of the immaterial soul on the other (OCM I 215, LO 101–2). But all the modes of the soul are, in their own nature, independent of the body – a disembodied soul could enjoy the same phenomenology as an embodied one.20 Malebranche’s other great debt to Descartes is to his mechanistic natural philosophy. Throughout his career Malebranche would make modifications to Cartesian mechanism (regarding, for example, the laws of motion), but he never dreamed of abandoning the vision that inspired the whole programme. The basic idea is brilliantly simple, and is set out with striking lucidity in Book Six of the Recherche (OCM II 321–45, LO 453–66). Starting with the clear and distinct idea of matter as essentially just three-dimensional extension, one can derive all its possible modes (shapes, motions and arrangements of parts). Ideally, one would then 6
INTRODUCTION
go on to derive the laws of succession of these modes, i.e. the laws of motion. (Malebranche would eventually concede that this could not be done a priori.) The ‘substantial forms’ and ‘real qualities’ of the scholastics are dismissed as unintelligible verbiage. As for the (supposed) souls of animals, they too are eliminated: Malebranche is one of the firmest advocates of the notorious Cartesian doctrine of the bête-machine.21 Natural philosophy shows, he argues, that a beast–machine is possible, i.e. that a sufficiently complex machine could perform all the actions of animals; metaphysics and theology provide positive reasons for thinking the doctrine is true. A just God, for example, could not permit innocent suffering; but the beasts are innocent, so they do not suffer. Malebranche’s debts to Augustine are equally profound. At the heart of Augustine’s philosophy lies the simple thought that nothing of value can be created or sustained without the direct involvement of God. This ‘theocentrism’ or ‘supernaturalism’22 has striking implications in fields as diverse as ethics, natural philosophy and epistemology. In ethics, the Augustinian emphasises the dependence of our virtues on God’s gift of grace and the impossibility of our developing any genuine virtues without this divine aid. (The picture painted by the Stoics of their sage reeks of pride and folly.) In natural philosophy, the exquisitely contrived forms of plants and animals testify not only to transcendent divine design but also to immanent divine guidance in the form of ‘seminal reasons’ dispersed by the Creator throughout His creation. Likewise in epistemology we err if we think we can obtain any knowledge by our own powers. Just as the eye needs light to see, so the mind requires ideas in order to know. Malebranche takes both the analogy and the moral from Augustine.23 As the sun is to the eye, so is God to the human mind, he explains in Book Six of the Recherche (OCM II 244–8, LO 408– 10). There is, he writes to Arnauld, no such thing as natural knowledge (OCM VI 144). We commonly contrast ‘natural’ with ‘revealed’ knowledge, but the distinction we should be drawing is between two different types of revelation. This ‘illuminationist’ epistemology is Augustine’s main legacy to Malebranche. Human knowledge, Malebranche will argue, is only possible at all if our ideas are God’s ideas, i.e. if we have epistemic access to the archetypes God uses in His ongoing creation of our world. This is one of a number of topics on which Malebranche uses Augustine to correct Descartes (OCM XVIII 51). Descartes generally writes as if he thinks of ideas as merely modes of our own souls – this is certainly how Arnauld, like most second-generation Cartesians, reads him. But if this were the case, Malebranche argues, there would be no justification for inferring ‘X is F’ from ‘My idea of X represents it as F’, 7
INTRODUCTION
and hence no escape from veil-of-ideas scepticism (OCM IX 915). But God’s Idea of X is the archetype in accordance with which X was created, so it is logically impossible for it to misrepresent X. If my ideas are God’s ideas, objective knowledge is secured. Only this ‘Vision in God’, counterintuitive and visionary as it may seem, blocks scepticism by licensing inferences from the properties of things-as-represented by their ideas to the properties of the things themselves (OCM XII 7–26, JS 3– 18). On two other issues of crucial importance Malebranche also parts company with Descartes. He accepts neither Descartes’ voluntarism about necessary truths nor his rejection of final causes in natural philosophy. Descartes tells us that God could have made it the case that 2 + 2 = 5, although of course we cannot conceive how this could be.24 Given Malebranche’s Augustinian conception of ideas, he must dismiss this Cartesian thesis as absurd: for 2 + 2 to make 5 would require a change in the eternal and immutable intellect of God Himself, which is nonsense. God’s mere will, Malebranche argues in Éclaircissement X (OCM III 127–61, LO 612–32) does not of itself constitute Order; rather, it is subject to an Order which exists eternally in the divine intellect. As for final causes, Descartes explicitly rejects them in his Principles of Philosophy, on the ground that we are completely ignorant of God’s designs and purposes.25 But such false modesty, replies Malebranche, has dangerous consequences for morality and religion. If we are completely ignorant of God’s purposes, we must abandon the traditional argument to design. A plenum of nauseous excrement, or a single atom in an endless void, would prove God’s wisdom as well as our rich, diverse, and harmonious cosmos. Final causes, Malebranche insists in the correspondence with Arnauld, are useless in physics but essential in theology (OCM VIII 674). If we really believed ourselves ignorant of God’s purposes in creation, we could find no evidence in nature for His wisdom or benevolence.26 Yet another important departure from Descartes concerns our knowledge of our own souls. When I reflect on my own thoughts and feelings, says Malebranche in Éclaircissement XI (OCM III 163–71, LO 633–8), I become aware of myself as a thinking and feeling being. When I think about matter, I grasp its essence by means of my clear and distinct idea of extension. I can show, by means of a negative argument a priori, that thinking and feeling cannot be modes of any material thing. But I still lack a clear and distinct idea of the soul. Doubtless there is such an archetype of the human soul, a model or blueprint rendering intelligible the substance and its diverse modes, but this idea exists only 8
INTRODUCTION
in the mind of God and has not been revealed to humans. I have therefore only a confused and obscure knowledge of the modes of my own soul: I have no rational insight into its essence. There is no rational psychology, no science of the soul in the strict sense (i.e. a body of demonstrated knowledge) which stands to the soul as geometry does to matter.27 Malebranche’s modifications of Augustine’s philosophy are less radical. Here he sees himself as simply bringing Augustine up to date with developments in natural philosophy in general, and with the mechanical philosophy in particular. This point is made in the preface of the Recherche (OCM I 20, LO xxxix–xl) and in Éclaircissement XV (OCM III 235–6, LO 676). Augustine, for example, accepted that colours, tastes, smells, etc. were properties of bodies, but a short tutorial on the mechanical philosophy and on the subjectivity of such sensible qualities would have been sufficient to make him change his mind. And Augustine never doubted the common belief in animal souls, but this may simply indicate a failure on his part to reflect deeply enough to rid himself of a vulgar prejudice. The Cartesian thesis of the bête-machine should have appealed to him on metaphysical and theological grounds, had he ever become aware of it. Nothing essential to Augustine’s central philosophical concerns, Malebranche insists, is inconsistent with a mechanistic natural philosophy.
Faith and reason For Malebranche, as for Augustine, all knowledge is revelation. Revealed knowledge, however, is of two distinct kinds. Our clear ideas are revelations of God’s ideas, which are the models or archetypes employed in creation. Rational knowledge (in mathematics, metaphysics and ethics) is therefore a kind of natural revelation, available to all men who are willing to reflect and to meditate. By contrast, there is the revelation of God’s Word in scripture, as interpreted over the centuries by the authorities of the Church. Since these two kinds of knowledge stem from the same source, the divine logos, they cannot contradict one another. Any apparent contradictions between human reason (philosophy) and Christian faith (dogmatic theology) must be illusory, the products of misunderstanding. In the Traité de la Nature et de la Grâce, for example, Malebranche warns his readers against interpreting Saint Paul as a Calvinist (OCM V 63). To suppose that God has predestined the great majority of mankind to eternal damnation would be to make scripture contradict reason, which tells us clearly that a perfectly wise and benevolent being could have no such sinister design.
9
INTRODUCTION
But if faith and reason cannot contradict one another, two distinct possibilities remain. They may give rise to entirely independent disciplines of philosophy and theology, each with its own separate domain. Or they may provide one another with mutual support and illumination, giving rise to a coherent body of philosophical theology. Descartes had taken the former view. In the Meditations he attempted, of course, to prove the existence of God and the real distinction of mind and body, but he regarded these as philosophical theses, supported by philosophical arguments. As for dogmatic theology (the Trinity, the Incarnation, the Eucharist, Heaven and Hell) Descartes refused – at least in his published works – to meddle with such sensitive matters.28 For this consistent and sustained separation of philosophy and theology he was praised by Arnauld,29 who carefully divided his own objections to the Meditations into ‘philosophical’ and ‘theological’. But this vision of two independent disciplines with distinct and clearly demarcated domains was alien to Malebranche. When Arnauld, following his usual practice, divided his objections to the Traité into ‘philosophical’ and ‘theological’ points, Malebranche protested against being thus cut in two. ‘One must not say that I act sometimes as a philosopher and sometimes as a theologian: for I always speak, or claim to speak, as a rational theologian’ (OCM VIII 632). The first significant critic of the Recherche, the sceptic Simon Foucher, accused Malebranche of confusing matters of faith with matters of reason. Our author, says Foucher in his critique, first takes care to distinguish the mysteries of faith from the clear ideas of reason, and then mixes them up again! To discuss the fall and the incarnation in a work of philosophy is, in Foucher’s eyes, a serious error: such matters should be left for sermons.30 A similar objection would later be made by Fontenelle against Malebranche’s Traité de Morale.31 In an appendix added to later editions of the Recherche, Malebranche replied to Foucher’s objection (OCM II 493). To combine, he explains, is not to confound. To speak of God, for example, according to our idea of an absolutely perfect being belongs to philosophy, which can prove what dogmatic theology must assume, i.e. that God is not a deceiver. And, as we shall see, there may be many aspects of the world of our experience that philosophy can only discover, but theology can explain. The crucial point is that philosophy does not rely upon theology for its premises. In the whole of the Recherche, Malebranche insists, Foucher ‘will not find that I suppose any article of faith as a principle …’ (OCM II 495). What Malebranche is seeking to do is to re-establish the harmony between reason and faith which he saw breaking down around him. Scholastic Aristotelianism, in his view, had failed. For all the subtleties 10
INTRODUCTION
of the Jesuits, the naturalism of Aristotle’s philosophy could never be assimilated into a Christian world view. What was needed was a new and genuinely Christian philosophy.32 In this new philosophy, there will be a subtle interdependence of reason and faith, more or less along the following lines. Reason supporting the foundations of faith
Christianity presupposes the existence of a veridical God: dogmatic theology would be impossible without the assumption that the Word of the Lord, as contained in scripture, is true. But, says Malebranche in Book Six of the Recherche (OCM II 372, LO 482) and again in the Conversations Chrétiennes (OCM IV 14), the dogmatic theologian cannot prove this assumption. If proof is demanded, it must come from the philosopher. Likewise, the Christian doctrine of the afterlife requires the assumption that conscious experience does not end with bodily death. The Cartesian philosopher, unlike his Aristotelian colleague, can support the doctrine of the afterlife by demonstrating the real distinction of mind and body. This proof of the real distinction, says Malebranche in a letter of 1693, does not amount to a demonstration of immortality, but does establish a powerful presupposition in its favour. Since the passage from nothingness to being is incomprehensible to the human mind, the immortality of the soul has been sufficiently demonstrated to philosophers when one has proved that the soul is a substance distinct from the body, for by nature it is only the modes of beings that perish (OCM XIX 605). In seeking to show, by philosophical argument, the existence of a veridical God and an immortal soul, Malebranche was not being adventurous or innovative: such proofs were required of Catholic philosophers by the Council of Trent.33 In attempting to use Cartesian philosophy to explain Christian theology, however, Malebranche would be venturing into much more perilous territory. Reason explaining the mysteries of faith.
It has always been permitted, Malebranche insists in the Traité (OCM V 7–8), and again in his replies to Arnauld (OCM VIII 632), to use philosophy to explain the mysteries of the Christian religion. This is not a matter of seeking to impose new dogmas: innovation in theology is 11
INTRODUCTION
always a mark of error. One may nevertheless offer new accounts or explanations of such dogmas as the Trinity, the Incarnation and the Eucharist. This has always been the practice of the Church: as Augustine was to Plato, or Aquinas to Aristotle, so Malebranche proposes to be to Descartes. It is necessary, he says, ‘to make philosophy serve theology’ (OCM VIII 643). In this way, he hopes, we can ‘silence the unbelievers, and … place reason on the side of religion’ (OCM VII 592). In proposing new accounts of the Christian mysteries, Malebranche’s aim is not merely to explain but also to justify the ways of God to man. Only thus, he argues, will we ‘faire taire les impies’. Pierre Bayle, for example, in his famous article Manichaeism from the Dictionnaire, had argued that the disorders and evils of our world are better accounted for by the Manichaean doctrine of two opposed deities, one good and the other evil, than by the orthodox Christian belief in a single wise, omnipotent and benevolent God.34 Malebranche seeks to answer Bayle (and the libertins in general) by his account of the creation, the fall and the incarnation. Unfortunately, his attempts to defend particular Christian dogmas by philosophical argument are by far the weakest part of his entire oeuvre. And once one admits the competence of reason in this domain, one runs the risk of arriving at manifestly heretical conclusions.35 A just God, the freethinkers began to argue, would not (could not) act in the unjust and arbitrary manner described by Christian dogma. Faith explaining the difficulties of reason
In examining both the external world and the workings of the human mind, we encounter what seem to be manifest design flaws. The world contains deserted wastes, natural disasters, plagues and famines – on the face of it, there seems no reason to believe that it was created for us by a wise, powerful and benevolent God. Even our minds show traces of bad design. How often do we deliberate carefully about what to do next, form a considered judgement ‘X is better than Y’, and then proceed to choose Y rather than X? Our choices should be determined by our reflective judgements of value, yet all too often this is not the case in practice. In Book Seven of the Nicomachean Ethics Aristotle described this phenomenon, and gave it the name of akrasia (weakness of will, lack of selfcontrol), but to describe is not to explain, and the existence of akrasia remained an anomaly for the rationalistic ethical theories of the Greeks. Here, says Malebranche, philosophy stumbles over puzzling limits to its comprehension. Knowing that the natural world is the creation of an all-perfect Being, we can’t understand why it should be so manifestly flawed. And knowing that our reflective and considered value judgements 12
INTRODUCTION
should determine our choices, we are at a loss to explain why they so often fail to do so. Here faith comes to the aid of reason, and theology illuminates philosophy. According to the Christian doctrine of the fall, both the natural world and the human soul are mere ‘ruins’, imperfect remains of a perfect prior creation.36 The moral is drawn by Malebranche in the second Éclaircissement of the Traité (OCM V 182). The Christian philosopher, he there claims, can shed light on problems that would forever vex the mere natural philosopher. In the fourth dialogue of the Entretiens, Ariste asks his teacher Theodore how it can be that, contrary to all the principles of Order, the superior being (the human soul) often finds itself subordinated to and dependent on the inferior (the human body), and thus distracted from its true good (union with God). Theodore replies as follows: ‘Yes, no doubt, there is some mystery there. My dear Aristes, how philosophers are indebted to religion, for only religion can extricate them from the predicament in which they find themselves!’ (OCM XII 101, JS 63). If natural philosophy and psychology fall into hopeless perplexities and insoluble problems which only faith – in this case, the Christian doctrine of the fall – can resolve, then we have the makings of a powerful inference to best explanation for the truth of the religious dogmas. Critics like Foucher, who accused Malebranche of subordinating philosophy to theology, have missed the point here. We don’t need any reliance on faith to see that the natural world and the human soul are both badly flawed creatures; this is evident from observation and a little reflection. Likewise, we don’t need to rely on faith to prove the existence of an all-perfect God. But then we find ourselves, as mere philosophers, wondering how these two truths can be compatible. We need recourse to faith, Malebranche thinks, if we are to explain how a perfect Creator could make an imperfect universe. Malebranche is not basing his philosophy on theology in the objectionable sense that he derives philosophical conclusions from theological premises. Rather, the facts to be explained are already known, independent of Christian dogma, and the argument takes the unobjectionable form of an inference to best explanation of those facts. If Malebranche is right, natural philosophers and psychologists will find themselves drawn into Christianity in the attempt to complete their own projects.
Difficulties The Cartesian method of doubt, Malebranche warns his readers early in the Recherche, does not extend to matters of faith. In philosophy, it is wrong to assent to any proposition without proof; merely plausible 13
INTRODUCTION
reasons are never good enough. In theology, however, the opposite is the case: here it is the demand for evidence that is a dangerous source of heresies and errors. ‘In a word, to be among the Faithful, it is necessary to believe blindly; but to be a philosopher, it is necessary to see with evidence …’ (OCM I 62, LO 14). In Book Four, Malebranche contrasts those who believe blindly whatever they are told with those who demand evidence in all things (OCM II 33–8, LO 278–82). The former people are just stupid, feeble intellects; the latter tend to become heretics or freethinkers. To be a good Christian is to accept the articles of faith in the absence of proof. But this submission to the authority of the Church on points of dogma does not extend to philosophical explanations of dogma. We must believe, for example, that the body of Christ is present in the wafer; we need not accept incomprehensible scholastic accounts of how this takes place. The dogma is from God, and must be accepted; to subject it to doubt would be un-Christian. The scholastic explanation of the dogma, however, is merely human; to doubt it is permissible and may even be required of us. In the Méditations Chrétiennes, this point is put in the mouth of the Word of God itself. ‘Know therefore’, we are told, ‘that evidence and faith can never deceive: but do not take plausibility for evidence, nor the opinion of some doctors for faith’ (OCM X 29). We have, according to Malebranche, one infallible guide, the divine word or logos, which speaks to us in two ways: by the inner light of reason and by the external light of revelation. But how does he deal with the sceptical problem of the criterion?37 Do we always know, from introspection, when a proposition is rationally self-evident? Or are we sometimes deceived into confusing mere plausibility with genuine selfevidence? And do we always know, with regard to supposed revelations, which books are genuine repositories of the Word of God, and how to interpret them?38 In answer to the first of these two sceptical questions, Malebranche has quite a lot to say. In the Méditations Chrétiennes, for example, he boldly puts his own metaphysics and epistemology into the mouth of the Word, but does so with the following warning in the preface. The Word, he says, is universal reason itself, and thus the source of all truth. I interrogate it, and try to record its answers faithfully. But ‘I have a great fear of not rendering its responses as I have received them; and even of not always distinguishing them from my prejudices, or from those obscure and confused sentiments inspired by the senses, the imagination, and the passions’ (OCM X 7). The same problem emerges in Éclaircissement XIII of the Recherche, which discusses the need for 14
INTRODUCTION
spiritual directors (OCM III 181–95, LO 644–52). When faced with a difficult decision, says Malebranche, one should withdraw into solitude and meditate: it is better to trust the inner light than any human guide. But we don’t always know whether we have effectively stilled our passions, and thus sometimes mistakenly think something evident when in fact it is not so. For this reason, spiritual directors may still on occasion be of use to us. In the Traité de Morale, the worrying implications of this admission are explicitly spelled out. Sometimes, Malebranche admits, it is hard to be certain that one is listening to the Word, because ‘the secret promptings of the passions are of the same nature as this inner sentiment. For when one acts against opinion and custom, one often feels inner reproaches very like those of reason and Order’ (OCM XI 68). But if conscience can torment us for a mere violation of some arbitrary social custom, mistaken for an edict of Order itself, Malebranche has failed to resolve the epistemological problem posed by the sceptics. They demanded a criterion of truth; Malebranche, it now appears, has failed to provide one. If we rely on introspection and phenomenology alone, we may fall into error, yet nothing better is on offer. What is delivered by the inner light is self-evident and infallibly true. But not everything that seems self-evident to introspection is the genuine deliverance of the inner light.39 In applying the problem of the criterion to revealed knowledge, Malebranche shows himself at his most dogmatic and conservative. The question is raised in Entretien XIII, but Malebranche’s spokesman Theodore, in launching a bitter attack on Protestantism, does little but parrot Catholic commonplaces. The universal Church is, he insists, ‘the blessed storehouse of tradition’ (OCM XII 324, JS 257). It is ‘perpetual through all time and universal in all places’ (OCM XII 326–7, JS 259). But to identify the doctrine of ‘the universal Church’ with that of late seventeenth-century Roman Catholicism is simply question-begging. Many Protestants claimed that their doctrines were those of the early Church, and that the theology of latter-day Roman Catholicism was little more than a mass of corruption. Malebranche never seriously addresses this problem. The dogmas of the Church include some that are explicitly labelled and taught as ‘mysteries’. Such doctrines, Malebranche insists, must be accepted as matters of faith even in the absence of clear ideas. In Book Three of the Recherche he denounces those theologians who seek to rid Christianity of its mysteries (OCM I 390–6, LO 203–6). The Socianians cannot understand the Trinity and the Incarnation, and are thus led to 15
INTRODUCTION
deny that God could become man; the Calvinists cannot understand how the body of Jesus Christ could be present in the communion wafer, and thus deny that it is (literally) present at all; some philosophers cannot reconcile God’s omniscience with human liberty, and end up denying free will. Even orthodox theologians sometimes fall into such errors. Here, for a moment, Malebranche sounds positively antirationalistic: It might be said … that the objections raised against the main articles of our faith, especially against the mystery of the Trinity, are so strong that they cannot be given solutions that are clear and convincing and that do not in any way shock our feeble reason, for these mysteries are indeed incomprehensible. (OCM I 395, LO 206) This passage is, however, uncharacteristic of Malebranche’s considered views. Elsewhere, and especially in his later works,40 he attacks antirationalism in theology. In the Traité de Morale, for example, he denounces the claim that blind obedience is the principal virtue of Christians. Such an attitude, he warns, only flatters the pride of ecclesiastical superiors and justifies the sloth of inferiors; it is unworthy of the true Christian (OCM XI 36). We must, he advises, accept the dogmas of the faith without doubts or reservations, but we are always permitted to seek further enlightenment regarding them. Given that the true philosophy of nature (Cartesianism) tells us that the essence of matter is extension, for example, we may legitimately enquire how this is compatible with the Christian dogma of the Eucharist. In an unpublished letter, Descartes had broached this sensitive subject, only to drop it when it threatened to lead him into dangerously controversial theological territory.41 Malebranche was prepared to go further, developing, in the context of a vigorous polemical exchange with the Jesuit Louis le Valois, a properly Cartesian account of the real presence (OCM XVII–1 445ff). The details need not concern us here: the crucial point is that the search for intelligibility in theology is not forbidden a priori. What is prohibited is the inference from failure to falsehood, i.e. from ‘We can’t make sense of p’ to ‘It cannot be true Christian doctrine that p’. But is the very search for clear and distinct ideas implicitly antiChristian, in the spirit if not in the letter? Is the intellectual independence of the Cartesian philosopher already subversive of religion? Many Christians of the period thought so, and would have endorsed Bossuet’s public attack on Malebranche’s theology as excessively rationalistic: 16
INTRODUCTION
I see, not only with regard to this point about nature and grace, but also with regard to many other very important articles of religion, a great conflict brewing against the Church under the name of the Cartesian philosophy.42 Malebranche, it should be clear, is walking a fine line, trying to remain within the Catholic Church while developing an epistemology of religion that is closer to the Quaker theory of the ‘inner light’. Problems arise when we try to explain a Catholic dogma but fail miserably in the attempt or (as so often with Malebranche himself) fall into blatant excesses of special pleading. To maintain orthodoxy, we must never permit the inference from ‘I can’t understand how p could be true’ to ‘p cannot be true’. An obvious example would be the inference from ‘I can’t understand why a just God should care whether someone sprinkles water on a baby’s head’ to ‘a just God wouldn’t care about something as trivial and morally insignificant’. Malebranche himself will have nothing to do with such reasoning. But when the gulf between our moral and metaphysical understanding on the one hand, and the dogmas of religion on the other, is filled by only feeble reasoning and manifest special pleading, doubts will become pressing. Bad explanations of how p could be true may backfire and cast doubt on the proposition itself. The reader may find himself thinking ‘p could be true only if q were true, but q is incredible’. Instead of asking how a wise and just God could permit the eternal damnation of most men for a sin committed by a remote ancestor, one may find oneself concluding that this simply could not be the case at all. Since reason, for Malebranche, is universal, the reason of the Christian is no different from that of a pagan or even an atheist.43 By his attempts to rationalise Christian doctrines, Malebranche is prepared to meet unbelievers and sceptics on their own ground. As Paul Hazard says, Malebranche’s valiant attempts to show the reasonableness of the Christian religion may have served, in the end, only to undermine it.44 The conservative Bossuet may have been right after all.
17
T E N S I O N S I N C A R T E S I A N M E TA P H Y S I C S
2 TENSIONS IN CARTESIAN M E TA P H Y S I C S
Richard Watson’s Downfall of Cartesianism In his 1966 book The Downfall of Cartesianism,1 and in a succession of later books and articles,2 Richard Watson has painted a wonderfully vivid portrait of the trials and tribulations of the rationalist metaphysics of the seventeenth century. Although his central claim was scarcely original – he ascribes it himself to the seventeenth-century sceptic Simon Foucher, whose critique of Cartesianism he largely follows – the brilliant lucidity of his exegesis and the light his approach throws on major aspects of post-Cartesian metaphysics combine to make the book a seminal work in late twentieth-century history of philosophy. Watson’s central claim was that Cartesian metaphysics collapsed because it failed to meet its own standards or norms of intelligibility. The Cartesians, he claimed, were committed to two key principles, which he labelled the Causal Likeness Principle (CLP) and the Epistemological Likeness Principle (ELP). According to CLP, all causal relations are subject to the principle that the effect must resemble the cause, so that the causal relation is a sort of transformation of the patient into the form or likeness of the agent. The begetting of a horse by a horse, or the heating of a cold body by a hotter one might serve as examples of such a process. According to ELP, an idea in the mind of a knower must represent its object by way of resemblance, i.e. by sharing certain properties with it. The obvious analogy would be with the way in which a picture can represent its object. Cartesian metaphysics, according to Watson, was committed to these principles by its very rationalism, by the quest for intelligibility that lies at its heart. They belonged to the ‘hard core’ of the research programme, and were therefore non-negotiable. But Cartesian metaphysics was also committed to a variety of claims about what substances exist and how they interact, and these claims seem flatly incompatible with CLP and
18
T E N S I O N S I N C A R T E S I A N M E TA P H Y S I C S
ELP. Descartes’ strict dualism of mind and body provides the most obvious and familiar source of such problems. If the impact of a stream of corpuscles on my retina can cause a visual sensation in my immaterial mind, what becomes of CLP? How can there be resemblance between cause and effect in such a case? Likewise for the role of the will in voluntary bodily motions: where is the resemblance between the volition in my mind and the motion of my legs? As for ELP, a corporeal image imprinted on the brain might resemble a physical object, but Descartes consistently denied that his ideas were corporeal images. They are, he says, not images, but ‘like’ images in the way they represent objects. This is scarcely enlightening. A non-extended thing can’t represent an extended thing by way of resemblance. Watson’s book has proved a wonderful source of insight for students, young and old, of seventeenth-century metaphysics. Once the student has grasped CLP and ELP, and seen how the Cartesians were both committed to them and to metaphysical theses that violate them, the big picture is suddenly illuminated. Given this understanding of the tensions that drove late seventeenth-century metaphysics, we could better understand both the transformations of Cartesianism effected by Spinoza, Malebranche and Leibniz, and the eventual collapse of traditional metaphysics in the face of sceptical attacks. Although I am, as should be obvious, a great admirer of Watson’s work, I don’t think he got it exactly right. In particular, I think his emphasis on resemblance is over-played. The rationalist metaphysician is, by the nature of his project, committed to certain a priori principles of intelligibility governing the causal and epistemological relations. He is, let us say, committed to causal and epistemological intelligibility principles, CIP and EIP. But resemblance, although it may be the most obvious, is not the only way in which intelligibility can be secured. The idea of a circle, for example, need not itself be a circle. A circle can be made known to the mind by way of an image (resemblance), or by way of its mathematical formula (a² + b² = r²), or by way of a mechanical rule of construction. We shall see the relevance of this when we come to discuss Malebranche on étendue intelligible or ‘intelligible extension’. The best way to understand seventeenth-century rationalist metaphysics, I suggest, is to see it as firmly committed to principles of causal and epistemological intelligibility, CIP and EIP, and then engaging in fierce controversy over how CIP and EIP are to be interpreted. Does intelligibility require resemblance, in which case my CIP and EIP would be indistinguishable from Watson’s CLP and ELP? Or might causal and epistemological intelligibility be secured by relations other than 19
T E N S I O N S I N C A R T E S I A N M E TA P H Y S I C S
resemblance? We have already seen how EIP might be satisfied without resemblance. Does the same hold for CIP? Descartes suggests in Meditation Three that the cause must be adequate to produce the effect, i.e. that it must have enough ‘reality’ or ‘perfection’ to do the job. If it did not, then some or all of the reality or perfection of the effect would have been produced ex nihilo, which is impossible. But this causal adequacy principle (CAP) is perfectly consistent, he thinks, with causal interaction between mind and body, i.e. without resemblance. And CAP also allows ‘eminent’ causality, where the cause of an effect with property F is not actually F, but possesses the reality or perfection of F in some higher form. What I propose, then, is that we replace Watson’s CLP and ELP with my CIP and EIP. For the moment these can be stated in somewhat vague terms, since a central object of controversy among second-generation Cartesians is precisely how these principles should be stated, what they require and what they exclude. CIP: For a genuine causal relation to hold between cause C and effect E, there must be some intelligible connection between C and E. This notion is usually glossed in terms of what could in principle be discerned by rational insight or nous by a sufficiently enlightened mind. Such a mind, we are told, could ‘see’ the effect in the cause, and thus perceive a priori that the effect would result from the cause. The most obvious but not necessarily the only way in which CIP could be met is by some sort of ‘containment’ of the effect ‘in’ the cause. EIP: For a mind M to know the properties of an object O by way of its idea I, the idea must be ‘present’ to the mind, and the presence of I to M must render intelligible to M (some of) the properties of O. The Aristotelians would say that the form of the object, without its matter, must be present in the knowing mind. The Cartesians reject the Aristotelian terminology, but must find some adequate replacement that preserves the same basic thought. Once again, the most obvious – though not, perhaps, the only – explanation would be in terms of literal resemblance, which would assimilate ideas to images. Why is it important to distinguish my CIP and EIP from Watson’s CLP and ELP? The answer will become clear in subsequent chapters where we discuss Malebranche’s views about causality (occasionalism) and about ideas (the Vision in God). Malebranche’s work lies at the heart of the Cartesian metaphysics discussed in Watson’s book, which focuses on the period from 1673 to 1712. But Malebranche, in seeking to defend
20
T E N S I O N S I N C A R T E S I A N M E TA P H Y S I C S
rational principles of intelligibility, is led to deny both of Watson’s resemblance principles, at least as they would naively be understood. If causal and epistemological intelligibility can only be secured by resemblance then Malebranche’s position is flatly self-contradictory. Charity thus requires us to regard Cartesian metaphysicians as committed to the somewhat looser and more open-ended principles I have spelled out. We can then see whether there is a consistent position that affirms CIP and EIP, but denies CLP and ELP. If we assume from the start that rationalist metaphysicians are absolutely committed to resemblance principles, it is too easy to convict them of inconsistency. Our strategy in this chapter is as follows. We shall look briefly at Descartes’ views on the nature of ideas (the second section), then spend some time on the accounts of ideas in later Cartesians such as La Forge, Arnauld and Régis (the third section). After that, we shall turn our attention to Descartes’ views about causation (the fourth section) and the nature and extent of his concessions to occasionalism. The final section (the fifth section) discusses the views of later Cartesians such as La Forge and Cordemoy on causation, tracing the emergence of occasionalist themes and arguments. The chapter provides essential background for our reading of Malebranche. Without a proper grounding in the intellectual context provided by Descartes and by second generation Cartesian metaphysics, the writings of Malebranche would be almost incomprehensible.
Descartes on ideas Descartes has a great deal to say about ideas, but still leaves many questions unanswered. What exactly are Cartesian ideas, how do they arise, and how do they serve their function in the cognition of objects? His readers must search through scattered remarks in his published works and his correspondence, seeking as ever the most plausible and coherent overall interpretation of his views. The following four quite different kinds of things are labelled ‘ideas’, at various places in the Cartesian corpus: 1 2 3 4
Corporeal images in the brain. Direct or immediate objects of thought or perception. Mental acts or operations. Mental dispositions.
The first supposed sense of ‘idea’, as a corporeal image traced on the brain, can be set aside. Descartes does use ‘idea’ in this sense in early
21
T E N S I O N S I N C A R T E S I A N M E TA P H Y S I C S
works such as the Regulae (AT X 414 and 441, CSM I, 43 and 56–7), and the Treatise on Man (AT XI 174–7, CSM I, 105–6), but he never uses the term in this sense in the Meditations, and explicitly distinguishes ideas from corporeal images in a number of the Replies. In the Second Replies, for example, he writes as follows: I do not myself think that the idea is of the same kind as the images of material things which are pictured in the imagination; I maintain it is simply that which we perceive with the intellect, when the intellect apprehends, or judges, or reasons. (AT IX 109, CSM II 99) There are corporeal images in the imagination, but they must be sharply distinguished from ideas proper: Thus it is not only the images depicted in the imagination which I call ‘ideas’. Indeed, in so far as these images are in the corporeal imagination, that is, are depicted in some part of the brain, I do not call them ‘ideas’ at all; I call them ideas only in so far as they give form to the mind itself, when it is directed towards that part of the brain. (AT IX 124, CSM II 113) In the Third Replies, he accuses his critic (Hobbes) of confusing ideas with images in his criticisms of the Third Meditation: Here my critic wants the term ‘idea’ to be taken to refer simply to the images of material things which are depicted in the corporeal imagination; and if this is granted, it is easy for him to prove that there can be no proper idea of an angel or of God. But I made it quite clear in several places throughout the book, and in this passage in particular, that I am taking the word ‘idea’ to refer to whatever is immediately perceived by the mind. (AT IX 141, CSM II 127) In a letter to Mersenne of July 1641, the same point is reiterated: For by ‘idea’ I do not just mean the images depicted in the imagination; indeed, in so far as these images are in the corporeal imagination, I do not use that term for them at all. (AT III 392–3, CSMK III 185)
22
T E N S I O N S I N C A R T E S I A N M E TA P H Y S I C S
In his post-Meditations writings, Descartes takes pains to draw a sharp distinction between corporeal images traced on the brain, and ideas in the strict sense, which lie on the mental side of the mind–body divide. Mental functions such as imagination and memory may depend on corporeal images,3 but the idea should never be identified with the image. Unfortunately, Descartes continues to say that ideas are ‘like’ (comme) images, which is unhelpful. An image can represent a body by resemblance; an idea, it seems, cannot do so in any straightforward sense. Having set aside ideas in sense 1 (corporeal images), we can go on to set aside – as irrelevant to our current concerns – ideas in sense 4 (dispositions). In the 1641 letter to Hyperaspistes and in his Comments on a Certain Broadsheet (written against some theses of his wayward erstwhile disciple Regius) Descartes seeks to spell out his conception of innate ideas, and thus to prevent possible misunderstandings of innatism. I have no doubt, Descartes writes to Hyperaspistes, that the soul of the human infant concerns itself with the demands of the body rather than with lofty metaphysical subjects: None the less, it has in itself the ideas of God, of itself and of all such truths as are called self-evident, in the same way as adult human beings have these ideas when they are not attending to them; for it does not acquire these ideas later on, as it grows older. I have no doubt that if it were released from the prison of the body, it would find them within itself. (AT III 424, CSMK III 190) The fact that some men – and even perhaps whole peoples – never think explicitly about God does not refute the Cartesian claim that the idea of God is innate: I do not doubt that everyone has within himself an implicit idea of God, that is to say, an aptitude to perceive it explicitly; but I am not surprised that not everyone is aware that he has it or notices that he has it. (AT III 430, CSMK III 194) In the Comments on a Certain Broadsheet, Descartes is concerned not only to defend innatism, but also to explain what the innatist is and is not committed to. He dismisses out of hand the crudest interpretation of innate ideas: ‘I have never written or taken the view that the mind
23
T E N S I O N S I N C A R T E S I A N M E TA P H Y S I C S
requires innate ideas which are something distinct from its own faculty of thinking’ (AT VIII 357, CSM I 303). To say that a given idea is innate is to say that the mind has the power to form it from its own resources, without requiring instruction from without. Ideas may thus be said to be innate in the sense that … we say that generosity is ‘innate’ in certain families, or that certain diseases such as gout or stones are innate in others: it is not so much that the babies of such families suffer from these diseases in their mother’s womb, but simply that they are born with a certain ‘faculty’ or tendency to contract them. (AT VIII 358, CSM I, 303–4) If ideas in the dispositional sense (4) are merely capacities or tendencies to have ideas in some more robust sense (2 or 3), then we can regard them as being called ‘ideas’ only in some secondary or parasitic sense. Let us set them aside, and concentrate our attention on the central sense or senses of ‘idea’ in Descartes. There is no doubt whatsoever that Descartes often – indeed, standardly – uses ‘idea’ in sense (3), to stand for mental acts or operations, which are modes or modifications of the immaterial substance of the soul. All ideas, he tells us in Meditation Three, are ‘modes of thought’ (modi cogitandi). As such, they differ from one another not in their ‘formal’ or intrinsic reality but in their ‘objective’ (representational) reality (AT IX 31, CSM II 27). The same point is reiterated at Principles, 1, 17: When we reflect further on the ideas that we have within us, we see that some of them, in so far as they are merely modes of thinking, do not differ much from one another; but in so far as one idea represents one thing and another represents another, they differ widely. (AT IX 32, CSM I 198) There is no doubt that this use of ‘idea’ for mental act or operation is fundamental for Descartes. In his well-known and helpful article, Vere Chappell refers to such mental acts as ideas(m), or ideas taken ‘in the material sense’.4 But an idea can always be thought of in two different ways, or under two different aspects. We can think of it either in terms of its intrinsic features, as an act or operation of thinking substance, or we can think of it in terms of its role in cognition, i.e. in terms of its representative content. If we think of ideas in this latter sense, we may
24
T E N S I O N S I N C A R T E S I A N M E TA P H Y S I C S
find ourselves positing what Chappell calls ideas(o), ideas taken ‘in the objective sense’. Ideas in this sense are differentiated by their contents, which allow one person to entertain the same idea (in this sense) on several different occasions, or for two or more people to have the same idea. It is ideas(o) that Descartes is gesturing towards when, in the Second Replies, he defines ‘idea’ as the form of a thought: II. Idea. I understand this term to mean the form of any given thought, immediate perception of which makes me aware of the thought. Hence, whenever I express something in words, and understand what I am saying, this very fact makes it certain that there is within me an idea of what is signified by the words in question. (AT IX 124, CSM II 113) We thus need to talk of ideas(o) if we assume that thoughts can be shared and communicated. We may also need to posit ideas(o) as immediate objects of perception, if we are to make sense of non-veridical perceptions. Descartes does speak in a number of places, e.g. in the First Replies, of ideas as immediate objects of perception. But are Chappell’s ideas(o) a distinct set of entities, intermediaries between the mind and the world of bodies? This traditional interpretation of Descartes has been coming under increasing fire in recent years.5 Chappell suggests that an idea(m) and its idea(o) differ only by what Descartes would call a ‘distinction of reason’, i.e. they are the same thing conceived in two different ways, and could not exist separately.6 If this is correct, we will need to distinguish between two different interpretations of ideas as mental objects: (2a) Ideas = mental objects really distinct from ideas(3), capable of serving as the immediate objects of perception, and thus as intermediaries between the perceiving soul and external objects. (2b) Ideas = mental objects not really distinct from ideas(3), but rather the objective contents of ideas(3). We may still talk of ideas in this sense as ‘immediate objects’ of perception, but this only means that in any perception the mind has a direct grasp of the content of its thought. As we shall see, the second generation Cartesians, led by Arnauld,7 interpreted Descartes as committed to ideas(3), acts of the soul, and ideas(2b), the contents of ideas(3), but denied that Descartes’ theory
25
T E N S I O N S I N C A R T E S I A N M E TA P H Y S I C S
had any need for ideas as distinct intermediate objects (2a). An idea, Arnauld will explain, can always be seen under two different aspects or in two different relations: qua mode of the soul it modifies it can be called ‘perception’; qua (purported) representation of some external object it can be called ‘idea’. But one should not infer from this duality of aspects or relations a real duality of things. Arnauld takes this view to be faithful to Descartes, and we have seen no reason to doubt him. Insofar as Descartes had a clear conception of the nature of ideas, it may well have been this ‘double aspect’ theory subsequently articulated by Arnauld. Before leaving Descartes, we must ask his views on the source of our ideas, since this issue would become one of fundamental importance for the later Cartesians and for Malebranche in particular. In the Third Meditation Descartes divided ideas into three categories: innate, adventitious (sensory) and artificial. The idea of God is innate (stamped on the human soul at its creation, like the craftsman’s trademark on his work); the idea of a dog is adventitious (derived from sense experience); the idea of a centaur is artificial (made up by the Greek poets). The reader might be permitted to think that our sensory ideas are directly caused by the physical stimuli affecting our sense organs. In the Comments on a Certain Broadsheet, however, Descartes rejects any such concession to empiricism. In an important sense, he now insists, all our ideas are innate: … if we bear well in mind the scope of our senses and what it is exactly that reaches our faculty of thinking by way of them, we must admit that in no case are the ideas of things presented to us by the senses just as we form them in our thinking. So much so that there is nothing in our ideas which is not innate to the mind or to the faculty of thinking. (AT VIII 358, CSM I 304) Our minds form the ideas of colours, tastes, smells, etc. on the occasion (Descartes’ word) of certain corporeal motions in our sense organs and the sensory centres of our brains, although there is no resemblance between those motions and the resulting idea. The lack of any resemblance between our sensations of, e.g. heat and colour and any properties of external objects is of course a cornerstone of Descartes’ mechanistic account of sense perception, explicitly emphasised in Le Monde and the Dioptrique. Why, asks Descartes at the start of Le Monde, ‘should nature not also have established some sign which would make us have the sensation of light, even if the sign contained nothing in itself which is similar to this sensation?’ (AT XI 4, CSM I 81). There may be
26
T E N S I O N S I N C A R T E S I A N M E TA P H Y S I C S
picture-like images imprinted on the brain, he writes in the Dioptrique, but ‘we must not think that it is by means of this resemblance that the picture causes our sensory perception of these objects – as if there were yet other eyes within our brain with which we could perceive it’ (AT VI 130, CSM I 167). The sensation of red does not resemble the spin of the corpuscles of the second element; the sensation of heat does not resemble the agitation of the corpuscles of a hot body. Should we then take these sensations as non-intentional, devoid of any representative content? Or should we think of them as confused perceptions8 ? Malebranche will take the former view, Arnauld and most of the later Cartesians will take the latter. Strictly speaking, Descartes concludes, none of our ideas may be said to be drawn from, or copied from, the impressions made by external objects on our sense organs. Even where there is resemblance between the brain trace and some properties of the external object (in respect of size, shape and arrangement of parts), the brain trace or image can’t itself get into the mind. The body, it seems, can prod or prompt the mind; it can’t inform or instruct it. But if our sensory ideas are not drawn from objects, how do they come about? Three possible accounts come to mind: a b c
Sensory ideas are produced by the mind itself, on the occasion of a physical stimulus. Sensory ideas are directly and individually produced by God, on the occasion of a physical stimulus. All the sensory ideas a given human mind will ever need are placed in it by God in advance, in its creation. The role of the physical stimulus, on this account, is merely to evoke something already present.
Descartes’ language in the Comments strongly suggests theory (a). God, according to this theory, has created the human mind with a certain range of active powers, prominent among which is the power to produce its own sensory ideas on the occasion of the physical stimulation of the sense organs of its body. But this theory ascribes an astonishing range of powers to the mind, none of which it is aware of possessing or exercising. As Malebranche would later insist, my sensations may be ‘en moi’ but they also take place ‘sans moi’ and even ‘malgré moi’, i.e. against my will. This might lead us to reject theory (a) in favour of (b) or (c). In his eliminative argument for the Vision in God Malebranche will explicitly reject all three of the above theories.
27
T E N S I O N S I N C A R T E S I A N M E TA P H Y S I C S
Later Cartesian accounts of ideas Among the disciples of Descartes, a rough consensus gradually emerged about the nature of ideas. The fundamental sense of ‘idea’, it appears, is idea(m), ideas taken in the material sense as acts or operations of the soul, and therefore as modes of spiritual substance. In this sense, there is no distinction between ideas and perceptions. If we need on occasion to speak of ideas as immediate objects of thought or perception, and thus to posit ideas(o), we should be on our guard not to think of these as having any really distinct existence. In having a thought or a perception we are immediately aware of its content, and can thus speak of this content as the object of the thought or the perception. This allows us to make sense of cases where there is nothing to think of (Descartes’ malicious demon) or to perceive (the alcoholic’s pink rats). Louis de la Forge expresses this view clearly in his Traité de L’Esprit de L’Homme. Ideas, he says, can be considered in two ways, either in themselves or in their representative role. When considered in themselves, they can be defined as ‘the forms, modes or ways of thinking of the mind by the immediate perception of which we perceive the thing which they represent to us and are certain that we have such a thought’.9 I call them ‘modes’ or ‘ways’ of thinking because I cannot conceive of any of them without a substance which thinks, although I can very easily conceive of this substance without any one of them in particular. Thus they are not distinct from the substance of the mind, nor are they distinct from each other except as different modes of the substance to which they belong.10 La Forge also quotes a lengthy passage from Descartes’ Comments on a Certain Broadsheet, and endorses Descartes’ claim about the origin of our ideas.11 Material things, La Forge notes, can only transmit motions to the sense organs of our bodies; these motions cannot reach the spiritual soul: … these corporeal species could only be modes or accidents of some body which have these two properties: first, they are incapable of leaving their subject to go into another one; and secondly, even if they could leave, they could be received only into an extended subject. It follows clearly from this that they could not be received into our mind. Therefore since all our kinds of knowledge are acts which do not leave the inside of 28
T E N S I O N S I N C A R T E S I A N M E TA P H Y S I C S
the soul, it is impossible for corporeal species to be the form by the immediate perception of which we have such or such thoughts.12 What La Forge calls ‘corporeal species’ are patterns of activity in our brains and nervous systems. Such patterns, he tells us, are not in general iconic, and should not be thought of as little images imprinted on the brain. But even if they were images, it would make no difference, because all that these ‘species’ can impart is motion, and motion cannot be imparted to the immaterial substance of the soul. The role of these ‘species’ is simply to prompt the soul’s own faculty of forming ideas. If asked about the causes of our ideas, La Forge explains, we need to distinguish the ‘primary and effective’ cause from the ‘remote and occasional’ one. A body can only ever be the remote and occasional cause of an idea, something which ‘by the union of mind and body, causes our faculty to think and determines it to produce the ideas of which the faculty of thinking itself is the principal and effective cause’.13 All our ideas thus originate from the mind as primary and effective cause; some – those labelled ‘adventitious’ by Descartes – also require the remote and occasional cause of a physical stimulus. This account, says La Forge, allows us to answer the famous question whether our ideas are innate or acquired: I reply that they are both. They are innate in the mind not only because the mind never received them from the senses but also because it is created with the faculty of thinking and of forming ideas and is the principal and proximate cause of them, in the same way in which one says that gout and stones are innate in certain families if those born into them carry dispositions towards those illnesses. But ideas are acquired and are not innate if by this word ‘innate’ is meant that they are in the substance of the soul as in a reservoir, in the way one arranges pictures in a gallery to look at them when one wishes. For there is none of them in particular which needs to be actually present in our mind which, since it is a substance which thinks, cannot have anything actually present without being aware of it. That is why they are contained in the mind only in potency and not in act, in approximately the way in which shapes are contained in a piece of wax. There is this difference, however, that in the wax this potency is merely passive, whereas in the case of the mind it is also active.14
29
T E N S I O N S I N C A R T E S I A N M E TA P H Y S I C S
But this last difference, merely noted in passing by La Forge, is of fundamental importance. A piece of wax has the passive power of plasticity, of allowing itself to be shaped by external pressures. To explain the shape it has taken on, one cites the external forces that most recently acted on it and imparted a particular form. Such a passive power seems easily comprehensible. But the mind’s capacity to form ideas is not passive but active. Without being informed by the impressions on the senses, and without any resemblance to merely corporeal (and hence ‘remote’) causes, the mind must impart form to itself, producing its ideas in a manner that has a reliable one-to-one correspondence with those remote corporeal causes. Of this mysterious productive process we have no conception whatsoever. We also find it hard to understand how the modes of a purely immaterial substance can represent bodies. La Forge is aware of the difficulty: I realize that I will not avoid being asked, in this context, how it is possible for spiritual things such as the ideas or forms of our thoughts to make us think about the body and its properties with which they have no similarity or resemblance. That is not without difficulty. However you should not doubt what I have just said, especially if you keep two things in mind. First, since the mind is a thing which thinks, its nature is necessarily such that it can represent everything to itself by its own thoughts. Secondly, our mind is like a part of the divine mind. Now this infinite mind which knows everything in itself would not be able to know bodies if it were impossible for its thought (which is completely spiritual) to represent bodies to itself.15 This passage takes us right into the heart of the controversy between Malebranche and Arnauld, which we shall discuss in Chapter Four. Of the two considerations adduced by La Forge, the first leads to Arnauld, the second to Malebranche. Arnauld says that the human mind thinks, i.e. that its modes are representations, because that is its nature. It is simply a brute fact, incapable of further explanation, that God has created such beings. Malebranche says that we can think only because we participate in the divine mind through the Vision in God. But we are getting ahead of our story. Part One of the Port Royal Logic, the celebrated Art de Penser of Arnauld and Nicole, deals with conception, the first action or operation of the mind, and hence with the nature of ideas. The term ‘idea’, says Arnauld in Chapter One, is so clear and so simple that it cannot be explained in terms
30
T E N S I O N S I N C A R T E S I A N M E TA P H Y S I C S
of other words.16 There is a need, however, to distinguish clearly between ideas and the corporeal images formed in the imagination. The idea of a figure of 1,000 sides is clearly distinct from the idea of a figure of 10,000 sides, though it would be easy to confuse the corresponding images.17 We have a perfectly clear idea of our own thoughts, though we form no image of thought. Affirmation and negation are clearly distinct mental operations, although the image (a round earth) might be exactly the same for the opposite judgements ‘The earth is round’ and ‘The earth is not round’. Arnauld’s conclusion is plain: When therefore we speak of ideas, we do not call by that name the images which are painted in the imagination, but everything which is in our mind when we can say truly that we conceive a thing, in whatever manner we conceive it.18 Ideas, he continues, are the bearers of the meanings of words, so whenever we understand a word such as ‘God’ or ‘soul’, it follows that we have a corresponding idea, even if we find ourselves unable to represent such things in the imagination. But where do our ideas come from? The empiricist position, most recently championed by Gassendi, holds that all our ideas are derived from the senses. Arnauld starts by denying that empiricism can account for such ideas as those of thinking and existence. By what sense, he asks, do we derive these ideas? Are they bright or dark, blue or red, treble or bass, hot or cold, bitter or sweet? From what sensible images, and by what processes of recombination, could such ideas have been formed? If the empiricist can come up with no remotely credible answers to such questions, then, argues Arnauld, … it must be admitted that the ideas of being and of thought do not in the least derive their origin from the senses, but our soul has in itself the faculty of forming them, although it often happens that it is excited to do so by something which strikes the senses …19 It is therefore false, Arnauld concludes, that all our ideas come from the senses. Indeed, one should go further, and deny that any of our ideas owes its origin to the senses: … it could be said, on the contrary, that no idea which is in our minds derives its origin from the senses, except by occasion, in
31
T E N S I O N S I N C A R T E S I A N M E TA P H Y S I C S
that the movements which take place in the brain, which is all that the senses can produce, give the occasion to the soul to form for itself various ideas which it would not form for itself without that, although almost always these ideas have no resemblance to what takes place in the senses and in the brain, and there are furthermore a very great number of ideas which, having no relation to any corporeal image cannot, without an evident absurdity, be referred to our senses.20 The Art de Penser does not discuss in any detail the precise ontological status of ideas. Arnauld’s views on the subject are, however, perfectly clear from his other works. He regards ideas as acts or operations of the soul, equivalent to perceptions. In Vraies et Fausses Idées, this is made perfectly explicit: ‘I also take the idea of an object and the perception of an object to be the same thing’.21 Although there is no real distinction between idea and perception, there is a difference of aspect. When we speak of it as a perception, we are thinking of its formal or intrinsic reality as a mode of the soul; when we speak of it as an idea we are thinking of its intentionality, its representational content: I have said that I take the perception and the idea to be the same thing. Nevertheless it must be noted that this thing, although only one, has two relations: one to the soul which it modifies, the other to the thing perceived insofar as it is objectively in the soul; and that the word perception indicates more directly the first relation and the word idea the second.22 If we ask Arnauld to explain how it can be that the modes of the human soul have such representational content, he will answer that it is their nature to do so, i.e. that that is simply how God created the human soul. Intentionality is accepted as the mark of the mental, and is regarded as irreducible, not susceptible to further explanation. This null theory of intentionality, treating it as a brute fact about the mind, was something that Malebranche, as we shall see, could not accept. To say that the modes of our soul represent external objects because it is their nature to do so is, he insists, to explain nothing at all. Another spokesman for this Cartesian orthodoxy was Pierre Sylvain Régis, whose Cours entier de Philosophie or Système général selon les principes de M. Descartes appeared in three volumes in 1690. Régis explicitly identifies ideas with perceptions, which are simply modes of the soul. By the terms ‘idea’ and ‘perception’, he writes, ‘I do not
32
T E N S I O N S I N C A R T E S I A N M E TA P H Y S I C S
understand anything besides the modes of thinking which I know by themselves; but which represent to me things which are external to myself’.23 Régis criticised Malebranche on a number of points in three letters dealing with the Moon Illusion, the nature of ideas, and the pleasures of the senses.24 Malebranche’s Réponse a M. Régis was included in later editions of the Recherche. The second part of the reply, dealing with the nature of ideas (OCM XVII–1 279–310) reveals clearly that Régis followed closely the account of ideas we have found in Arnauld. When asked by Malebranche whether the idea of infinity can itself be finite, Régis replies by invoking Descartes’ distinction between ‘formal’ (intrinsic) and ‘objective’ (representational) reality. Qua mode of a finite spiritual substance, i.e. considered in its intrinsic nature or formal reality, the idea of infinity (or of God) is finite; qua representation, i.e. considered in its ‘objective’ reality, it is infinite (OCM XVII–1 285).25 A similar stratagem copes with the generality of ideas. Each idea, qua mode of a particular soul, is itself a particular, but qua representation it can be general in its signification (OCM XVII–1 302). On Régis’ view, according to Malebranche, the soul is itself a monde intelligible, i.e. it can know all things merely by considering its own perfections (OCM XVII–1 291). As we shall see in the next chapter, Malebranche’s argument for the Vision in God includes a sustained attack on this notion of the soul as a monde intelligible in its own right, a theory attacked by Malebranche on both philosophical and theological grounds.
Descartes on causation Descartes’ considered views about second causes and natural powers continue to perplex commentators and to provide plenty of material for lively scholarly debate.26 At first sight, it would seem that he is committed to the full range of natural causal powers recognised by common sense. Four such powers will be central to our discussion. They are the power of a moving body (1) to set another body in motion, or (2) to cause sensations in a mind; and the power of a mind (3) to move its own body by an act of will, and (4) to excite ideas in its own substance. But there are three central themes in Cartesian metaphysics that seem to tell against the acceptance of natural causes and to impel Descartes towards occasionalism. In the first place, there is the strict substance–mode metaphysic. Substances are capable (DV) of independent existence; modes are only conceivable as modifications of substances. So if motion is a mode of the moving body, it becomes impossible to conceive how it could be
33
T E N S I O N S I N C A R T E S I A N M E TA P H Y S I C S
transferred from one body to another. In a letter to Henry More of 1649, Descartes distinguishes bare motion from motive force, and suggests that the moving force of a body in motion may not be in the body itself but in God, the ultimate cause of all bodily motions. ‘The transfer which I call “motion” is no less something existent than shape is: it is a mode in a body. The power causing motion may be the power of God himself preserving the same amount of transfer in matter as He put in it in the first moment of creation’ (AT V 403–4, CSMK III 381). So motion, which is a mode of the moving body, cannot be transferred; and motive force or power is not truly in the moving body at all.27 The problem is even worse for the action of minds on bodies. If I can’t understand how one body can push another, how will I be able understand how a volition (a mode of an unextended spiritual substance) can be the cause of the motion of an arm or a leg (a mode of extended corporeal substance)? When the substance–mode metaphysic is married to a rationalist conception of the causal relation, the problems become still more acute. But what thesis or theses of causal intelligibility is Descartes committed to? In the Third Meditation he is careful only to commit himself to a fairly weak thesis of causal adequacy. This causal adequacy principle (CAP) states only that the complete efficient cause of a given thing must contain at least as much reality or perfection as the thing it produces (AT IX 32, CSM II 28). If not, something would be coming from nothing. This principle, he insists, in response to the objections of Gassendi, is perfectly consistent with mind–body interaction. When asked how the soul can move the body if it is in no way itself material, and how it can receive the forms of corporeal objects, Descartes responds brusquely that the question betrays a fundamental misunderstanding: … the whole problem contained in such questions arises simply from a supposition that is false and cannot in any way be proved, namely that, if the soul and body are two substances whose nature is different, this prevents them from being able to act on each other. (AT IX 213, CSM II 275) This defence will hold if cause and effect are permitted to be equally real or perfect, and if the modes of finite minds and of bodies are supposed to be equal in their degree of perfection. Some recent commentators have taken Descartes at his word here and dismissed the supposed ‘scandal’ of mind–body interaction as a pseudo-problem.28 The difficulty with such a view is that Descartes appears to commit himself to stronger
34
T E N S I O N S I N C A R T E S I A N M E TA P H Y S I C S
theses of causal intelligibility than CAP.29 At a number of places, he seems committed to the claim that the cause–effect relation is perspicuous in the sense that a sufficiently enlightened mind could ‘see’ (i.e. rationally intuit) the effect in the cause. This suggests causal containment or preexistence rather than mere causal adequacy. In the Second Replies, for example, we read that ‘Whatever reality or perfection there is in a thing is present either formally or eminently in its first and adequate cause’ (AT IX 128, CSM II 116). There is nothing in the effect ‘which was not previously present in the cause, either in a similar or in a higher form’ (AT IX 106, CSM II 97). In the 1648 Conversation with Burman he tells his young interlocutor that ‘It is a common axiom and a true one that the effect is like the cause’ (AT V 156, CSMK III 339–40). This principle, he adds by way of qualification, applies only to the total cause, the ‘cause of being itself’. ‘Anything produced by this cause must necessarily be like it’. Even in the Third Meditation, as Radner has shown, one of the arguments turns on a causal containment principle rather than on the weaker principle of causal adequacy.30 There is therefore substantial evidence to suggest that Descartes endorsed a much stronger principle of causal intelligibility than CAP. Explicit or implicit at various points in his writings we can find a principle of causal containment or pre-existence to the effect that the effect preexists in the cause. Such a principle clearly rules out mind–body interaction. If modes cannot be conceived except as modifications of particular substances, then we can construct an even stronger argument for the conclusion that no mode can be communicated from any substance to another. This would rule out not just mind–body interaction but also body–body or mind–mind interactions.31 Maybe, as Leibniz was to suggest, the only intelligible causal relations are those between the modes of a single substance, unfolding in accordance with the formula that expresses its complete notion. The ‘windowless monads’ of the Monadology may represent the only metaphysical system consistent with the thorough working out of the implications of the substance–mode metaphysic. The third metaphysical theme drawing Descartes away from the naïve belief in natural powers and towards occasionalism is his belief that God’s conservation of His creatures is indistinguishable from their continuous (re)creation. In both the Meditations (AT IX 39, CSM II 33) and the Principles (AT IX 34, CSM I 200) Descartes writes that the existence of a creature at any given moment is logically independent of its existence at the previous moment. It follows that God must continually re-create each and every one of His creatures, none of which could exist for more than a moment without His sustaining action. The distinction between 35
T E N S I O N S I N C A R T E S I A N M E TA P H Y S I C S
preservation and creation is only a conceptual distinction, not a real one. In itself, the thesis of continuous creation was a perfectly orthodox piece of scholastic theology, found in Aquinas and many others. Now Aquinas and the majority of the schoolmen were firm believers in the reality of second causes. But in Malebranche, as we shall see, continuous creation is taken to imply the non-existence of any natural powers and to provide a rigorous proof of occasionalism. Two questions spring to mind. What has changed in the interpretation of continuous creation between Aquinas and Malebranche? And where does Descartes stand? To turn continuous creation from a commonplace of scholastic theology into a radical argument for occasionalism, additional theses are needed. Let us first state the most basic claim – the dependence of all created substances (material and immaterial alike) on God’s sustaining activity – as CC(S). CC(S) Every created substance (mind or body) depends for its continued existence on God’s sustaining action. What of the modes of created substances, e.g. the motions of bodies and the thoughts and volitions of finite minds? Are these also within the scope of continuous creation? The transformation of the doctrine between Aquinas and Malebranche turns on the following pair of theses, CC(AM) and CC(DM) which express, respectively, the continuous creation of all modes and of determinate modes. CC(AM) Every mode of every created substance S is as it is because of God’s re-creation of S, i.e. God re-creates not just S but all the modes of S. CC(DM) Every creature is re-created with a perfectly determinate set of modes. If God re-creates not just the substance but its modes, and if this recreation must maintain in existence a complete and perfectly determinate set of modes, then continuous creation entails occasionalism. The historical evidence suggests that this transformation of the doctrine of continuous creation occurred after Descartes, and that he would have been able to reconcile the weak version of continuous creation CC(S) with a belief in the reality of natural powers.32 If we re-examine Descartes’ views on the four types of causal relation outlined earlier, we find a mixed picture. When he sets out the rules of collision in the Principles, he writes cheerfully of the ‘force of motion’
36
T E N S I O N S I N C A R T E S I A N M E TA P H Y S I C S
and the ‘force of rest’, and begins to develop and articulate these notions into a worked-out system of dynamics.33 But when challenged about this ‘force of motion’, he is prepared to retreat in the direction of occasionalism. As he writes to Henry More, this ‘force’ may simply be God’s power, recreating bodies in such a way as to preserve the quantity of motion impressed on them in their creation.34 So the ‘force of motion’ of a moving body may need to be read in a sort of ‘as if’ sense. This might be interpreted as pure occasionalism, but it need not be. Descartes could be endorsing CC(S) but denying CC(DM), accepting that God’s re-creation is responsible for the continued existence of bodies, and for the conservation of the grand total of motion in the physical universe, but not for all the determinate motions of particular bodies. When it comes to the supposed power of bodies to cause sensations in minds, Descartes’ considered opinions are even harder to pin down. In the Comments on a Certain Broadsheet, as we have already seen, he suggests a sense in which all our ideas may be called innate. The mind, he suggests, has the power to produce ideas on the mere occasion of certain events in the sense organs and the brain. This theory, labelled ‘occasional causation’ by Nadler, is quite distinct from pure occasionalism in two important respects.35 In the first place, the true cause of our ideas is said to be the mind itself, not God. In the second place, the body is not deprived, on this theory, of all causal power. Even if the brain-state is not the primary or direct cause of the idea, it retains a prompting or triggering role. The mind has a sort of divinely preprogrammed innate repertoire, responding to physical stimuli arriving in coded form by producing representations, sometimes clear and distinct, sometimes confused, of the external world. So physical stimulus 17 elicits the idea of red and physical stimulus 18 elicits the idea of blue, without there being any perspicuous reason why this should be so. The physical stimulus is not the total cause, but still plays an essential role. Take this causal role away, and Descartes’ proof of the external world in the Sixth Meditation would collapse, leaving us with no defences against idealism.36 As regards the powers of finite minds, either to move their own bodies by acts of will, or to excite ideas or volitions in themselves, Descartes never seems to have entertained any doubts. ‘The freedom of the will is self-evident’, he tells us bluntly in the Principles (AT IX 41, CSM I 205–6). Supposing him to apply continuous creation in this domain, he must have supposed that God re-creates all the substances of finite spirits, and re-creates each with some thoughts or other, perhaps with a fixed ‘quantity of thought’ characteristic of each mind, but not with fully 37
T E N S I O N S I N C A R T E S I A N M E TA P H Y S I C S
determinate thoughts and volitions. He writes confidently in the Conversation with Burman that we know our own liberty on the basis of introspection (AT V 159, CSMK III 342), and never seems to doubt that this is an awareness of real causal powers, both over the voluntary motions of our bodies and over (many of) our thoughts. What should we conclude from this brief foray into Descartes’ views about causation? We seem almost to be engaged in an exercise in Hegelian dialectics. Thesis: Descartes accepts from common sense all four types of causal power: body–body, body–mind, mind–body, and mind–mind. Antithesis: Descartes endorses metaphysical theses (causal rationalism, continuous creation) that are strictly incompatible with naïve views about the existence of such causal powers. Synthesis: Descartes’ commitments to causal rationalism and to continuous creation are not of such a kind as to rule out acceptance of natural causal powers. Although God does continuously re-create each created substance, whether mind or body, He does not re-create it with a fully determinate set of modes. This in turn means that the explanation of any given mode of a particular mind or body can’t just be of the form ‘God willed it so’, but must refer to some secondary cause or natural agent. As for causal relations, it looks as if Descartes has to admit, as bona fide cases of ‘equivocal’ (but still genuine) causation, the power of a mind to move a body and of a body to prompt a mind to think.37 His considered view seems to be that the causal containment principle holds only for the total cause, which would allow us to continue to admit non-resembling causes.38 In giving everyday causal explanations, we hardly ever cite the total cause, generally citing only one salient causal factor among many, selected no doubt for pragmatic reasons. In his 1987 article on this topic, Dan Garber argued that Descartes conceived continuous creation as modally indeterminate, thus allowing some causal powers to both minds and bodies.39 In his 1993 discussion of the same issue, he changes his mind and concludes that Descartes denies causal powers to bodies but allows them to finite minds.40 This would make him a sort of semi-occasionalist. The authority of such a scholar should not be lightly dismissed. His later view also has the advantage that it fits a hierarchic conception of created substances, with the higher (mind) able to act on the lower (body) but not vice versa. I still incline, however, to the rival view of Scott 41 and Nadler 42 that Descartes is not an occasionalist at all, but holds views that allow him to reconcile a weak version of continuous creation with the reality of natural powers in minds and bodies alike. For a Cartesian, minds are more obviously loci of power, since each of us is immediately aware of such powers in 38
T E N S I O N S I N C A R T E S I A N M E TA P H Y S I C S
introspection, but it does not follow that bodies are altogether devoid of power. Bodies may contribute to the determination of the motions of other bodies, and may play a modest causal role in triggering the mind to produce ideas. The scholastic theory of divine concourse will be invoked to explain how the First Cause and a variety of second causes co-operate to produce any given effect.43 This section has discussed some of the recent scholarship on Descartes’ views about causation. It has provided grounds for thinking that Descartes was not as muddled about the subject as he is sometimes portrayed, and for resisting attempts to represent him as at least half way to occasionalism. My purposes, however, were relatively modest ones. I was not seeking to provide a perfectly consistent account of all Descartes’ utterances about causation. It may be that no such reading is possible. My main aim was to set up the tensions in Cartesian metaphysics that would trouble disciples such as La Forge and Cordemoy, whose writings would in turn set the stage for Malebranche.
Later Cartesian accounts of causation It is a commonplace in the history of philosophy that the second generation of Cartesians tended to embrace occasionalism. In the previous section we discussed some of the factors in Descartes’ metaphysics that were responsible for this transformation of Cartesianism.44 In this section we shall discuss the views of two of the most important of the intermediate figures between Descartes and Malebranche, Louis de la Forge (1632– 66) and Gerard Cordemoy (1626–84). We focus on these two secondgeneration Cartesians at the expense of others45 because of their influence on Malebranche – known in one case, inferred with overwhelming probability in the other. Malebranche refers his readers in Book One of the Recherche to Cordemoy’s Six Discours sur la Distinction et l’Union du Corps et de l’Âme, and expresses general approval of Cordemoy’s arguments for substance dualism (OCM I 123, LO 49). As for La Forge, we know that he contributed illustrations and critical comments to Clerselier’s famous edition of Descartes’ L’Homme, which made such an impression on the young Malebranche.46 We also know that a copy of La Forge’s own Traité de L’Esprit de l’Homme was found in Malebranche’s library. Some degree of influence of La Forge on Malebranche seems beyond reasonable doubt.47 Chapter 16 of La Forge’s Treatise discusses mind–body interaction. Given substance dualism, says La Forge, it is natural to worry about how mind and body can interact, and to suppose that it raises special difficulties
39
T E N S I O N S I N C A R T E S I A N M E TA P H Y S I C S
that do not arise for interactions between bodies. But this thought, however natural, is mistaken: I think most people would not believe me if I said that it is no more difficult to conceive how the human mind, without being extended, can move the body and how the body without being a spiritual thing can act on the mind, than to conceive how a body has the power to move itself and to communicate its motion to another body. Yet there is nothing more true.48 The natural error was vividly expressed by Lucretius, who taught that a body could be moved only by being touched, i.e. by the contact of another moving body. Lucretius assumed that this is the only intelligible manner of causal operation, but in this, La Forge insists, he was simply mistaken. To clarify our ideas we need to draw a distinction between motion itself, as it is in the moving body, and motive force or power. Considered in itself, motion is a mere mode of the moving body, its transfer from the vicinity of one body to that of others.49 It is nonsense to suppose that such a mode could be transferred from one body to another. The motive force or power of the moving body is, however, something quite distinct from this, and cannot be assumed to be a mode of the moving body at all. Every moving body, says La Forge, ‘must be pushed by something which is not itself a body and which is completely distinct from it’.50 How do we know that this motive force is not a property of any body? All we need to do is to consult our clear and distinct idea of material substance (three-dimensional extension) to see that the idea involves no such motive power. Bodies are moved, La Forge continues, by being successively re-created, in different places, by God. A body exists at place A because God creates it at A rather than at B. But, he continues, … not only can it not change its condition by its own power: I also claim that there is no creature, spiritual or corporeal, which can cause change in it or in any of its parts, in the second moment of their creation, if the Creator does not do so himself. Since it was He who produced this part of matter in place A, for example, not only must He continue to produce it if He wishes it to continue to exist but also, since He cannot create it everywhere or nowhere, he must put it in place B himself if He wishes it to be there. For if He put it anywhere else there is no force capable of removing it from that location.51
40
T E N S I O N S I N C A R T E S I A N M E TA P H Y S I C S
Even God cannot create a body without creating it somewhere. This looks like an explicit statement of the principle I earlier labelled CC(DM), continuous creation plus modal determinacy, at least for bodies. But this version of continuous creation, as Nadler reminds us, seems to establish too much.52 If developed in its full generality – i.e. applied to finite spirits as well as to bodies – it implies the complete non-existence of causal powers in the natural world. Malebranche would enthusiastically embrace that doctrine and seek ways of making its paradoxes more digestible; earlier Cartesians like La Forge would pull back from the brink. Although God is ‘the first, universal, and total cause’ of all the motions we observe in our cosmos, La Forge explains, ‘I also recognise bodies and minds as particular causes of these same motions …’.53 God has, for example, created human beings, establishing the union of mind and body: … it is easy to see that He who willed to join them in this way had to resolve at the same time to give to the mind the thoughts which we observe it acquiring on the occasion of motions of its body, and to determine the motions of its body in the way they should be in order to be subject to the mind’s will.54 This is unclear. Has God given the mind and body powers to act on one another in certain ways, or has He merely decided to intervene Himself? And is there any difference between body–mind and mind– body causation? It is worthy of note that La Forge uses the term ‘occasion’ for the first of these relations but not for the second. Maybe the mind has a genuine causal power over the body but not vice versa? La Forge’s considered view is never sufficiently articulated. He does however explicitly reject at least the language of full-blooded occasionalism: However you should not say that it is God who does everything and that the body and mind do not really act on each other. For if the body had not had such a movement, the mind would never have had such a thought, and if the mind had not had such a thought the body might also never have had such a movement.55 So La Forge rejects pure occasionalism, on the basis of a counterfactual thought-experiment. Unfortunately, the occasionalist could cheerfully endorse the test and its results. If the second cause (occasion) had not
41
T E N S I O N S I N C A R T E S I A N M E TA P H Y S I C S
been present, God would not have willed the effect, so it would not have taken place. So this test will not serve to distinguish believers in genuine natural powers from occasionalists. The soul has no power, La Forge thinks, to create or destroy motion, the total quantity of which is conserved by God. It can, however, redirect it, i.e. alter the direction of the currents of animal spirits in the pineal gland and the nerves.56 But this suggests that God’s re-creation of the physical universe is less than fully determinate, i.e. that He recreates each and every material substance, and re-creates the total ‘quantity of motion’, but leaves the determination of the modes of motion (which particles are moving in which directions and with which speeds) to second causes. So we have an unresolved tension between the strong interpretation of continuous creation as fully modally determinate (which La Forge seems to endorse) and his explanation of the human mind’s control over its own body. A similar problem will arise for the modes of finite minds, i.e. their thoughts and volitions. Once again, it seems, continuous re-creation must be less than fully modally determinate if the mind is to retain any genuine powers over its own thoughts and volitions. The active, self-determining power of the will is simply evident to consciousness, La Forge tells us (echoing Descartes); it is not something that can be rendered doubtful.57 But if the will is an ‘active principle’ by which the mind ‘chooses from itself and by itself and determines itself’58 then it surely follows that God, in sustaining the immaterial substance of each human mind, is not Himself the direct cause of each and every one of its modes. There is no evidence that any Cartesian before Malebranche embraced that radical opinion. When asked how mind and body can interact, given their completely different natures, La Forge has recourse to the scholastic distinction between univocal causes (where cause resembles effect) and equivocal causes (where this does not hold): It is obvious that the mind cannot act on the body as a univocal cause by forcing it to produce some thought and that the body likewise does not act on the mind by communicating some motion to it, because the mind cannot be moved nor can the body think. It must therefore be as an equivocal cause that the mind, by its thoughts, forces the body to move and that the body, by moving, provides an occasion for the mind to produce some thought. However it does not follow that the body is not the cause of the thoughts which arise in the mind on its occasion, nor that the latter is not equally the cause of the movements 42
T E N S I O N S I N C A R T E S I A N M E TA P H Y S I C S
which occur in the body as a result of its thoughts, just because they are only equivocal causes. For God is no less the creator of all things, and workmen are no less the authors of their workmanship, despite the fact that they are all merely the equivocal causes of these effects.59 Merely citing a familiar scholastic distinction, however, is not much help. If we want to know how spiritual modes and material modes can interact without any resemblance, it is not much help to be told that there is a familiar scholastic concept, ‘equivocal cause’ for such puzzling causal relations. Neither is the reference to God’s power helpful, since God was supposed to ‘contain’ all the perfections of His creation either ‘formally’ or ‘eminently’, i.e. in some higher form. La Forge, it seems, faces a choice between admitting the existence of utterly occult and unintelligible causal powers (i.e. abandoning causal rationalism) or retreating to occasionalism. Cordemoy’s Six Discours sur la Distinction et L’Union du Corps et de L’Âme (1666) takes us significantly closer to pure occasionalism. Cordemoy’s natural philosophy is fundamentally Cartesian, with the significant difference that he attempts to reconcile Descartes’ matter– theory with a version of atomism. After introducing his atoms in the first discourse, he proceeds in discourses two and three to set out a more or less orthodox mechanical philosophy of nature, with the local motion of subtle matter as the main cause (‘considering bodies alone’) of the phenomena.60 The explicit qualification is of course meant to prepare readers for the occasionalism of Discourse Four. Discourse Four is entitled ‘On the First Cause of Movement’, and begins with a proof that matter is not self-moving, and must therefore be set in motion by a non-material first mover. But the same cause (God) is needed to maintain bodies in motion as was required to set them in motion in the first place.61 It is this proposition, Cordemoy admits, which we find hardest to believe, however rigorously it has been demonstrated. We tend to assume that once a body has been set in motion it will continue in motion, and even that it possesses, by virtue of its motion, the power to move other bodies. But this is mere prejudice: When we say, for example, that the body B has driven the body C from its place; if we examine closely what we know for certain in this, we will see clearly that B was in motion, that it struck C, which was at rest; and that after this encounter, the first ceasing to be in motion, the second began to move. But when we think
43
T E N S I O N S I N C A R T E S I A N M E TA P H Y S I C S
that B gives movement to C, that is in truth only a prejudice, which stems from the fact that we only see, in this case, the two bodies; and that we are accustomed to attribute all the effects that are known to us to the things which we perceive, without taking care to note that often those things are incapable of producing such effects, and without considering that there could be a thousand causes which, however imperceptible they may be, could produce sensible effects.62 We can conceive of a body in motion without conceiving of any such motive power, Cordemoy continues. Indeed, we cannot conceive how such a power, assuming it to be (per impossibile) a mode of the moving body, could be transferred from one body to another. Given that the first mover of all bodies is a spirit, the only intelligible account of how motion is conserved and transferred from one body to another is occasionalism. The collisions between bodies thus provide occasions for a spirit (God) to re-allocate motions to bodies in accordance with some fixed set of rules. When we say that B has moved C we are therefore guilty, strictly speaking, of a misunderstanding, ‘citing the occasion in place of the cause’.63 What about our own bodily motions? A significant proportion of them are labelled ‘voluntary’ because they occur in accordance with our acts of will. Is the will the true cause of such voluntary motions? ‘No’, replies Cordemoy: to assume that would be to repeat the same error of confusing occasion with cause. Experience teaches that such motions often follow acts of will, not that our acts of will are their true causes. Strictly speaking, the volitions of finite spirits (human souls, angels, demons) are no more true causes of the motions of bodies than are collisions with other bodies. The deficiency of power of the human soul is simply a result of its finitude and dependency; only the will of God can be a true cause. The motions of all bodies are therefore both created and sustained by God’s power. He can, of course, give the direction of affairs to a subordinate, i.e. to an angel, but all that this means is that He can choose to direct His power in accordance with the angel’s will, not that the angel would become a true cause in its own right.64 In Discourse Five, Cordemoy goes on to apply this occasionalism to the mind–body union. What it means to say that a particular mind has been ‘united’ to a particular body is just that systematic relations have been established by God between the thoughts of one and the movements of the other. In creating human beings, God establishes these psychophysical laws, which He never breaks. Given such an account of 44
T E N S I O N S I N C A R T E S I A N M E TA P H Y S I C S
the causal relation, it is no harder to conceive the action of spirits on bodies, or bodies on spirits, than that of bodies on bodies.65 If we try to model mind–body causation on body–body causation, we find ourselves asking stupid questions such as how an incorporeal thing can exert a push, but such questions reveal only our own misconceptions. Everything acts according to its nature: spirits think and will; bodies move. Experience teaches that bodies move when struck by other bodies, and that our own arms and legs usually move at will, but the former type of causal action is no more intelligible than the latter. The only intelligible account of causation is occasionalism, and this copes equally easily with mind– body and body–mind causation as with body–body causation. If one read only the Six Discours, one might conclude that Cordemoy was indeed a complete and thoroughgoing occasionalist. It should be noted, however, that nowhere in the work does he discuss mind–mind causation, i.e. the power of the human mind to control (for the most part) its own thoughts. Nor does he explicitly link the doctrine of continuous creation with the thesis of complete modal determinacy, which would provide a clear demonstration that God is the direct cause of all my thoughts, feelings, and volitions. The Six Discours are not, however, our only source of evidence. In the second part of his Traité de Métaphysique Cordemoy addresses, with tantalising brevity, the question of human freedom.66 God, he explains, is the creator of everything real in our actions, without taking away our liberty. But does this make sense? One might suppose that God re-creates my soul at t2 with modes dependent on its modes at t1. But this seems to generate a regress, and still to leave no room for freedom. Perhaps God’s re-creation of the soul involves an element of modal indeterminacy? Such a theory could in principle allow space for human freedom. But was that in fact Cordemoy’s opinion? God, explains Cordemoy, incessantly propels each and every soul towards the good – this is the proper object of desire. But several alternative courses of action may appear to us as good. In such circumstances, we may experience indecision, and decide to suspend judgement. This suspension of judgement is an action on our part, as is the eventual decision to do X rather than Y. That decision could not of course be effective without God’s executive power, but the action remains ours rather than God’s. Two identical men, faced with the same pair of conflicting motives, might make different choices. In the light of such remarks, one might seek to save freedom by having God merely re-create souls with a desire for the good in general, i.e. with a sort of indeterminate yearning that becomes determinate only 45
T E N S I O N S I N C A R T E S I A N M E TA P H Y S I C S
as a result of our choices. Cordemoy’s illustration, however, suggests a very different account. He supposes two equally hungry men – let us call them Smith and Jones – who have the opportunity to take some food, and a moral reason (unspecified) for refraining. God impels Smith, by way of his appetite, to take the food and eat it, and impels Jones, by way of his conscience, to refrain. We are meant to suppose that both men experience both of these conflicting motives. This suggests that God doesn’t just create each soul with an indeterminate yearning for the good, but also creates it with a determinate set of motives. But if both men have both motives, what factor or factors determine which motive prevails? Does God give us conflicting desires without fixing their relative strengths? Or does our freedom reside merely in our power of suspending judgement? Do we, as it were, have an on–off switch within our control, so that we can prevent any given desire spilling over into precipitate action? Given vacillating motives, and such an on–off switch, one could account for the possibility of at least a measure of genuine freedom. Cordemoy never answers these questions. His attempt to sketch a sort of reconciling project, showing how God’s sustaining activity might be compatible with human freedom, remains too short and too vague to dispel our doubts and answer our queries. The questions he raises, however, are exactly those that would trouble Malebranche. When we come to Chapter Nine, we shall see Malebranche grappling – at much greater length and in much greater depth, if not, ultimately, with more success – with just this family of issues linking metaphysics, theology and morals. If we are to follow the argument from continuous creation to pure occasionalism, do we not inevitably find ourselves denying human freedom and making a nonsense of moral responsibility? Have we not provided every intelligent and well-informed sinner with the ultimate excuse: ‘it’s a fair cop, but God is to blame?’
46
THE VISION IN GOD
3 THE VISION IN GOD
The argument for ideas Malebranche’s most fundamental argument for ideas turns on his oftrepeated maxim, voir rien, ce n’est point voir. To see nothing is not to see at all.1 This axiom can be found in the Recherche (OCM II 99, LO 320), is repeated in the first of the Entretiens (OCM XII 35, JS 8), and reappears in the late Entretien d’un Philosophe Chrétien et d’un Philosophe Chinois (OCM XV 5). Malebranche assumes that any perception must have an object, i.e. that the proper analysis of perception is relational.2 But if perception always requires a really existing direct object, what are we to say about non-veridical sensory experiences? We could of course deny them the status of perceptions, but that would fly in the face of introspection. As far as phenomenology is concerned, dreams and hallucinations are indistinguishable from normal cases of sense perception. If we want to defend the voir rien principle and the relational analysis of perception that goes with it, we will find ourselves obliged to introduce ideas as intermediate objects, at least for cases of non-veridical perception. Everything we see clearly, directly and immediately, says Malebranche in Book Four of the Recherche, necessarily exists: I say what we immediately see, attest to, or conceive; for to speak strictly, the objects we immediately see are very different from those we see externally, or rather from those we think we see or look at; for in one sense it is true that we do not see these latter, since we can see, or rather believe we see, external objects that are not there, notwithstanding the fact that nothingness is not perceptible. But there is a contradiction in saying that we can immediately see what does not exist, for this is to say that at the same time we see and do not see, since to see nothing is not to see. (OCM II 99, LO 320) 47
THE VISION IN GOD
Since ideas are the objects of direct or immediate perception, they cannot be other than they appear. There is no reality–appearance distinction for ideas. They can therefore serve as bearers of phenomenal properties for non-veridical experiences. At a number of places, Malebranche trots out standard versions of the argument from illusion, taking such arguments to provide a definitive refutation of direct or ‘naïve’ realism about perception (OCM IV 73, OCM IX 945, OCM XV 5). To avoid contradiction, says Malebranche, we must carefully distinguish the direct and immediate objects of all perception (ideas) from the indirect and mediate objects of most sense perception (bodies). The voir rien principle applies to the former but not to the latter. Every perception has its direct and immediate object (an idea); most sensory perceptions also have an indirect and mediate object (a body). We do not perceive bodies immediately and in themselves, we learn in the third of the Conversations Chrétiennes, because we often see bodies that do not exist at all (OCM IV 61–2). A direct realist might respond at this point by adopting either of two different strategies. He could say that the direct object of, e.g. visual perception is usually a body but sometimes an idea. Or he could say that perception always takes a body (or at least an external object of some kind)3 as its direct object, and that some of the mental events we take to be perceptions are simply not perceptions at all.4 Neither of these responses would have impressed Malebranche. The Cartesian assumption of the transparency of the mental rules out the latter strategy. On Cartesian principles, I know with certainty when I am having a perception, and what I am (directly, immediately) perceiving. The possibility of error only arises when I infer from the idea that is directly present to my mind something about the external object that I naturally take to be its cause. As for the former strategy, Malebranche just seems to assume that all the objects of direct or immediate experience are of the same kind.5 So if some are ideas, all must be ideas. As for the possibility that bodies could be objects of immediate experience, Malebranche thinks it can be ruled out by a simple metaphysical argument. It isn’t just that we need to posit ideas to provide direct objects for dreams and hallucinations; every case of perception takes an idea as its direct object. The most famous (or infamous) argument for this thesis is the ‘walking soul’ of the Recherche. Everyone must agree, Malebranche tells his readers, ‘that we do not perceive objects external to us by themselves’.
48
THE VISION IN GOD
We see the sun, the stars, and an infinity of objects external to us; and it is not likely that the soul should leave the body to stroll about the heavens, as it were, in order to behold all these objects. Thus, it does not see them by themselves, and our mind’s immediate object when it sees the sun, for example, is not the sun, but something that is intimately joined to our soul, and this is what I call an idea. Thus, by the word idea, I mean here nothing other than the immediate object, or the object closest to the mind, when it perceives something, i.e. that which affects and modifies the mind with the perception it has of an object. (OCM I 413–14, LO 217) This much-derided passage6 makes it look as if mere spatial distance were what counted, but that cannot be Malebranche’s serious point. Nothing, after all, could be closer to me (spatially) than my own neurons, but I have even less awareness of these than of the sun and stars. In his reply to Arnauld’s objection, Malebranche describes the walking soul argument as ‘a sort of joke’ (OCM VI 95–6). The distance between my mind and the sun is not so much spatial as ontological and – more precisely – causal (OCM VI 212).7 The notion of ‘presence’, likewise, shifts from a literal (spatial) sense to a metaphorical (cognitive) one. The reason why the sun can’t be present to my mind is not where it is but what it is. Causation requires, on Cartesian principles, some intelligible relation between cause and effect, either resemblance or some analogue of resemblance.8 As we saw in Chapter Two, Cartesians such as La Forge and Cordemoy had already drawn the inference that bodies cannot, strictly speaking, act on our minds at all. The same moral could have been derived from Augustine, whose principles involve a natural hierarchy in which the higher (soul) can act on the lower (body) but not vice versa. The direct and unmediated perception of the sun by my soul would, Malebranche sees, violate both sets of principles. It would be a case of causation without intelligible connection, and it would involve a transgression of the ontological hierarchy. When I see the sun, Malebranche concludes, the direct and immediate object of my perception is an idea. Whatever ideas may turn out to be, they must be the sort of things that can act on the human soul. This, as we shall see, will provide Malebranche with one of his key premises in his argument for the Vision in God. But what are these ideas? They are, Theodore explains in the first Entretien, often overlooked by unreflective men. Ideas remain, however, real beings with an existence and properties 49
THE VISION IN GOD
of their own. Suppose, says Theodore to his disciple Ariste, that God were to annihilate all material things except our two bodies. Let us further suppose that God impresses on our brains all the same traces, or rather that He presents to our minds all the same ideas we have now. On this supposition, Aristes, in which world would we spend the day? Would it not be in an intelligible world? Now, take note, it is in that world that we exist and live, although the bodies we animate live and walk in another. It is that world which we contemplate, admire, and sense. But the world which we look at or consider in turning our head in all directions, is simply matter, which is invisible in itself and has none of those beauties we admire and sense in looking at it. (OCM XII 38, JS 10–11) In this passage, as elsewhere, Malebranche flirts with idealism by suggesting that bodies as such are actually invisible. When challenged on this point by Arnauld,9 however, he beats a tactical retreat. It does not follow from my principles, he explains, that we never see bodies, only that we do not see them ‘en eux-mêmes’ (OCM VI 101, OCM IX 959). We do see bodies, it now seems, but only indirectly, by means of the direct perception of their ideas. This indirect or representative realism has been the standard reading of Malebranche, from contemporaries such as Arnauld and Locke down to most of the modern commentators.10 In the final section of this chapter we shall discuss the radical reinterpretation that has been urged by one of the finest of modern scholars. Steven Nadler now thinks (revising his own earlier view)11 that Malebranche’s account of perception is a form of direct realism.12 Nadler’s challenge, I shall argue, requires us to clarify the sort of indirect or representative realism we ascribe to Malebranche, but does not force us to re-interpret him as a direct realist.
The eliminative argument for the Vision in God What is the ontological status of ideas? Cartesians like La Forge, Arnauld and Régis say that they are ‘modes’ or modifications of the substance of the soul, and add that they are, by their very nature, representations of (possible) external objects. It is here that Malebranche makes his most radical innovation in the fabric of Cartesian metaphysics. The ideas by which I perceive cannot, he contends, be modes of my own soul; rather, they must be archetypes of bodies in the all-seeing mind of God. The
50
THE VISION IN GOD
problem of intentionality, like the problem of causality, demands a supernatural solution.13 Malebranche’s argument for the Vision in God occupies chapters 1–6 of Part Two of Book Three of the Recherche (OCM I 413–47, LO 217–35), and is repeated, in all essentials, in the first of the Méditations Chrétiennes (OCM X 11–18). The form of his argument is somewhat surprising, at least at first sight. It is an eliminative argument, first listing the various hypotheses that might be thought to explain how the mind knows external objects, then seeking to eliminate all but the Vision in God. Of course this argument will only work, as Malebranche admits in his reply to the criticisms of Régis, if the list is exhaustive and the refutations definitive (OCM XVII–1 290–1).14 For the sake of logical completeness, we ought to list one further assumption, i.e. that the mind has knowledge of external objects at all. Without that assumption, the conclusion could only be disjunctive; either the Vision in God or scepticism.15 Malebranche is confident that his list is complete:16 We assert the absolute necessity, then, of the following: either (a) the ideas we have of bodies and of all other objects we do not perceive by themselves come from these bodies or objects; or (b) our soul has the power of producing these ideas; or (c) God has produced them in us while creating the soul or produces them every time we think about a given object; or (d) the soul has in itself all the perfections it sees in bodies; or else (e) the soul is joined to a completely perfect being that contains all intelligible perfections, or all the ideas of created things. (OCM I 417, LO 219) He sounds less confident, however, regarding the rigour of his eliminative proof of (e), which is his own hypothesis of the Vision in God: Perhaps we can resolve the question with some clarity though we do not pretend to give demonstrations that will seem incontrovertible to everyone; rather, we merely give proofs that will seem very persuasive to those who consider them carefully, for one would appear presumptuous were one to speak otherwise. (OCM I 417, LO 219) Let us examine Malebranche’s list of hypotheses, and then glance briefly at his refutations of the rivals to the Vision in God. Here is the list:
51
THE VISION IN GOD
H1
H2
H3a
H3b
H4
H5
External objects cause ideas in our minds by emitting their images or likenesses through the intervening medium. This is the Aristotelian theory of species. The soul creates its own ideas on the occasion of impressions being made on the sense organs by external objects. This is the production theory. God produced all the ideas a soul will ever need and placed them ‘in’ the soul at the moment of its creation. This is the storehouse theory of the soul. God produces in each soul just the ideas it needs on the occasions of particular patterns of stimulation of its sense organs. This is the occasionalist theory. The soul contains within itself all the perfections of other creatures, and can therefore know them ‘in’ itself. This is the monde intelligible theory. We see all things in God.
What is the source of this list? According to Connell, it is derived from scholastic accounts of angelic cognition, almost certainly by way of Suarez’s treatise De Angelis.17 The suggestion is not as far-fetched and implausible as it may seem. Although Malebranche is consistently scathing in his criticism of scholastic philosophy, hardly ever referring positively to scholastic sources, he would certainly have been familiar with mainstream authors such as Aquinas and Suarez. It is also clear that what Malebranche found objectionable in scholastic philosophy was its Aristotelian hylomorphism and its empiricism, and in particular its endorsement of the conception of the human soul as the form of the body. As a strict Cartesian dualist he could reject Aristotelian accounts of man and still draw on scholastic accounts of angels (purely immaterial substances) and angelic knowledge (completely independent of senses and sense organs) applying such accounts directly to the human soul and its cognitive powers. But even if Connell is right and the list was originally derived from Suarez, there can be no doubt that Malebranche thought it applicable to the second-generation Cartesian theories of La Forge, Arnauld, Cordemoy and Régis. Why might Malebranche think this division exhaustive? Is there some principled basis for it? The most plausible answer I know of comes from Connell18 and Nadler.19 H1–H4 can be divided, says Nadler, on the following basis. H1 and H2 are (roughly) empiricist doctrines; H3 and H4 are innatist, or at least non-empiricist. (Hypothesis 3b, the occasionalist theory, can scarcely be called a form of innatism, but it is clearly non52
THE VISION IN GOD
empiricist.) In H1 and H3 the mind is passive in receiving ideas; in H2 and H4 it is active. We therefore have the following division: H1
H2
H3a H3b
H4
Empiricist and passive. External objects simply impress their image or likeness on the mind, which is like wax passively receiving an impression. This is how Malebranche understands the Aristotelian species theory. Empiricist and active. The sense-impression is still necessary, but now serves as only the occasion for the mind to exert its native power and create for itself a corresponding idea. Innatist and passive. The mind is just a vast storehouse, stocked from its very creation by God with all the ideas it will ever require. Neither empiricist nor innatist, and passive. Ideas are neither derived from experience nor innate, but merely created by God as and when occasion demands. Innatist and active. The mind is itself an ‘intelligible world’ and as such is able to discover, by searching its own nature, the properties of external things.
Malebranche’s refutation of the species theory (OCM I 418–21, LO 220–1) is noteworthy only for its crude caricature of the theory under attack.20 He consistently confuses the Aristotelian theory with the Epicurean, construing ‘species’ as little material pictures emanating in all directions from the visible body and eventually striking our eyes.21 Malebranche rehearses a familiar battery of objections to this theory, mostly based on physical optics (perspective, interpenetration of rays, microscopes, wasting away of visible objects), but no Aristotelian would regard the critique as a fair one. Is Malebranche simply guilty of an elementary misunderstanding? Perhaps not. What he might say is that if one reasons only from clear ideas one must reject Aristotelian ‘forms’ and ‘qualities’, and accept that the physical world is purely mechanistic. On this basis he may feel entitled to state the species theory in a form that is at least intelligible, even if this intelligibility comes at the cost of easy refutation.22 In any event, H1 is dismissed without getting a fair and sympathetic hearing. What about H2, the production theory (OCM I 422–8, LO 222–5)? On this hypothesis, the external object makes an impression on the sense organs, which impression serves as the occasion for the mind to create an idea. Man thus shares in the work of creation. This theory is certainly in the offing in Descartes’ critique of Regius, and in La Forge’s commentary, and is ascribed by Malebranche to certain ‘Cartesian gentlemen’.23 This
53
THE VISION IN GOD
account, Malebranche retorts, elevates human powers too far. Ideas are real, spiritual beings, intrinsically nobler than bodies, so this view ascribes to the human soul the power to create something better than God’s creation. To suggest that the mind creates the idea from the material impression is no help at all – it is unintelligible how the latter could serve as raw material for the former. In any case, Malebranche continues, even if the soul could create its ideas, it would (like a painter) require prior familiarity with its object in order to produce a representation of it. A painter can’t depict an animal he has never seen; nor could the soul create an idea without an exemplar – i.e. without a prior idea. If one already possesses the idea, one doesn’t need to re-create it; if one doesn’t, one cannot create it ex nihilo for lack of a model. This theory thus ascribes to the human soul a miraculous power to create, and even to create with wisdom and order, yet without knowing what it is doing. As for H3a, the infinite storehouse theory of mind, it is hard to find anyone who actually held this theory.24 We have seen in Chapter Two that it was explicitly rejected by Descartes and La Forge as a misunderstanding of innatism. The hypothesis doesn’t occupy Malebranche for long (OCM I 429–32, LO 226–7). There are, he says, any number of types of geometrical figure, and infinite variety within each kind. There is an infinity of triangles, of quadrilaterals, of ellipses, and so on. We can think of any of these figures at will. Does it then follow that our minds must actually possess, prior to doing any geometry, ideas of them all? Is it reasonable to suppose that God, who always acts by the simplest means, should have furnished us with such a super-infinite stock of innate ideas? And if we did possess such a super-infinite stock, how would we ever find the things we needed in the infinite storehouse? Hypothesis H3b, the occasionalist theory, has sometimes been attributed to Cordemoy.25 It is briefly mentioned at the end of the chapter refuting the storehouse theory, as if it were merely a variant of that theory. But in fact it is a quite distinct hypothesis, and doesn’t involve God trying to cram an infinity of ideas into each and every human soul in its creation. So most of the arguments against the storehouse theory don’t work against the occasionalist theory. Malebranche sees the difficulty, and provides a new argument. The occasionalist theory, he tells us, is incompatible with our power to think of things at will. Furthermore, we must at all times actually have in us the ideas of all things, since we can at all times will to think about anything – which we could not do unless an infinite number of ideas 54
THE VISION IN GOD
were present to the mind; for after all, one cannot will to think about objects of which one has no idea. Furthermore, it is clear that the idea, or immediate object of our mind, when we think about limitless space, or a circle in general, or indeterminate being, is nothing created. For no created reality can be either infinite or even general, as is what we perceive in these cases. But all this will be seen more clearly in what follows. (OCM I 432, LO 227) At first sight, Malebranche’s criticism of the occasionalist theory looks inconsistent with his attack on the storehouse theory. He seems to be telling us, first, that the mind cannot have an infinite stock of ideas, and then, only a page later, that it must have an infinite stock of ideas. What he means, of course, is that the mind doesn’t have all these ideas actually present as part of its own resources, but that it must have access to an infinite stock of ideas if it is to be capable of intellectual activity. As he himself says, this will become clear once the Vision in God has been explained. What of H4, the theory that the mind contains within itself, ‘eminently’ (i.e. in a higher form) all the perfections of creatures, and is therefore a monde intelligible in its own right? This hypothesis, according to Malebranche, is the view of Arnauld26 and Régis (OCM XVII–1 291).27 Malebranche’s response takes us from metaphysics into theology (OCM I 433–6, LO 228–9). It is certain, he says, that the soul does see some things ‘in itself’, but this is true only of sensations such as pain and such supposed qualities as hot and cold, colours, tastes, etc. These are perceived by the soul in itself, without ideas, but are known in this direct way precisely because they are mere modes of the soul’s own substance, representing nothing at all beyond themselves. The claim made by some scholastics (and endorsed by Régis) that the human soul is itself a monde intelligible is dismissed by Malebranche as rash and presumptuous, a human attempt to usurp the privileges of God. All His creatures are indeed comprehended ‘in’ God: His intellect grasps each essence and perceives its possible instantiation; His will selects for actual existence some optimum subset of possible creatures. God therefore does know His creatures ‘in Himself’, i.e. He knows their essences in His intellect and their existence through His will. We, however, cannot claim to know either the essences or the real existence of things in a parallel manner.28 Having dismissed the rival theories as lacking any plausibility, Malebranche can go on to set out his positive view, the Vision in God. God, you will remember, re-creates all His creatures continuously, and 55
THE VISION IN GOD
does so in full awareness of what He is doing. It follows that there is something ‘in’ God that represents each creature. We also know that there is a peculiarly intimate union between the divine mind and human minds. Given these two things, the mind surely can see what in God represents created beings, since what in God represents created beings is very spiritual, intelligible, and present to the mind. Thus, the mind can see God’s works in Him, provided that God wills to reveal to it what in Him represents them. (OCM I 437, LO 230) God’s wisdom, says Malebranche, is manifest in the production of great effects by the simplest means. Since God can make a created mind see X simply by revealing to it His idea of X, He doesn’t have to pack an infinity of innate ideas into each and every created finite mind, as the storehouse theory H3a requires. Nor does He have to intervene in an ad hoc manner, as required by hypothesis H3b. His ideas are always at our disposal, so long as we ask the right questions in the right manner. Just as occasionalism emphasises the causal dependence of all creatures on the divine will, so the Vision in God emphasises our absolute cognitive dependence on the divine intellect. Just as I have no real causal powers of my own, but can only invoke His aid to move my own body, so I have no real cognitive powers of my own, but must invoke His aid in order to understand anything at all. In the final analysis, all knowledge is revelation, and attention to a problem (e.g. in mathematics) is just a form of prayer. Illumination always comes from without. Thus far Malebranche’s argument for the Vision in God has scarcely been overpowering. He has dismissed rival accounts of ideas as lacking in plausibility, and has ascribed to his own theory a couple of theological advantages, emphasising the wisdom of God (achieving the greatest ends by the simplest means) and our cognitive dependence on Him. Does he have any positive philosophical grounds for his theory? The strongest philosophical reason for accepting the Vision in God is, we are told, the following. When I want to think of an object I must, as it were, cast my mind’s eye over all objects – they must all be ‘present’ to me at once if I am to find what I am looking for. But this infinite capacity would not be possible except via the ‘union’ of my mind with the allencompassing mind of God: It is certain, and everyone knows this from experience, that when we want to think about some particular thing, we first glance
56
THE VISION IN GOD
over all beings and then apply ourselves to the consideration of the object we wish to think about. Now, it is indubitable that we could desire to see a particular object only if we had already seen it, though in a general and confused fashion. As a result of this, given that we can desire to see all things, now one, now another, it is certain that all beings are present to our mind; and it seems that all beings can be present to our mind only because God, i.e. He who includes all things in the simplicity of His being, is present to it. (OCM I 440–1, LO 232) We could present this as a transcendental argument in the Kantian manner. We all have a certain infinite capacity to think of whatever we choose. How is this capacity possible? Only, says Malebranche, by means of the Vision in God. Therefore the Vision in God is true. This is valid, but everything turns on the second premise, for which Malebranche provides no additional support. The Vision in God is also invoked by Malebranche to explain how the mind can possess general ideas, and how it can possess the idea of infinity in particular. Once again, we can see Malebranche as providing a quasi-Kantian argument from the existence of a science like pure mathematics to the Vision in God as its necessary condition.29 The argument is plainly Augustinian in its ancestry.30 But for Augustine the Vision in God is a theory of pure intellection only, an account of how we can know the eternal truths. Malebranche wants to extend the doctrine to the knowledge we possess, through sense perception, of changeable and corruptible things. God creates such things, so it follows that He must have knowledge of them. This knowledge need not, of course, resemble our sensory awareness. When I see a body like the sun, my perception consists of idée pure (a circular figure) and sentiment (yellow). God causes the sentiment in my soul but doesn’t share it; God discloses the idée pure which is ‘in’ Him as its archetype. What I see is the idea (the circle), but I find myself spontaneously ‘projecting’ the yellow (which is just a sensation) onto the object. I therefore see the object yellow-ly, as it were.31
The argument from properties Malebranche never abandoned the eliminative argument for the Vision in God. It is repeated in successive editions of the Recherche, reproduced in the Méditations Chrétiennes (1683), and explicitly defended in the
57
THE VISION IN GOD
Réponse à M. Régis (1693). He may vacillate regarding whether the eliminative argument provides a conclusive proof or merely a plausible argument, but he never retracts or qualifies his commitment to its validity. In his later writings, however, the eliminative argument for the Vision in God is supplemented (and, increasingly, supplanted) by a further argument which we can label the ‘argument from properties’32 . The target of the argument from properties is the account of ideas common to the mainstream second-generation Cartesians (La Forge, Arnauld, Régis). As we saw in Chapter Two, the emerging consensus among these Cartesians was that ideas should be identified with perceptions. Such mental acts are modifications or ‘modes’ of the substance of the soul. These modes are essentially representative; they are, by their very nature, representations of actual and possible objects. On this theory, intentionality is the mark of the mental, and is regarded as simply a brute and irreducible fact about the mental, not something susceptible to further explanation. In creating minds, God has created substances whose modes are representations – there is nothing more to say. Malebranche, not surprisingly, identifies this position with his H4, the monde intelligible of the eliminative argument. One way to regard the argument from properties is therefore as an ad hominem continuation of the eliminative argument. If it is agreed that the earlier hypotheses H1–H3 are untenable, then only H4 (the monde intelligible) and H5 (the Vision in God) are left standing. A refutation of H4 would then provide a proof – subject to the usual provisos of eliminative arguments in general – of H5. Malebranche’s argument against the identification of ideas with modes of the soul is quite straightforward in form. If one wants to prove that X≠Y, one must prove that X has some property that Y lacks, or vice versa. So if we list the properties of ideas, then the properties of modes of the soul, and find that the properties of ideas are quite distinct from and incompatible with the properties of modes of the soul, we have the desired proof of a real distinction. This argument can be found clearly spelled out in Book Four, Chapter Eleven, of the Recherche, in Éclaircissement X, and in the first of the Entretiens sur la Métaphysique. There are, says Malebranche, a number of ways of proving that ideas are distinct from perceptions: For it is clear that the soul’s modes are changeable but ideas are immutable; that its modes are particular, but ideas are universal and general to all intelligences; that its modes are contingent, but ideas are eternal and necessary; that its modes are obscure and shadowy, but ideas are very clear and luminous (i.e. its modes are 58
THE VISION IN GOD
only obscurely, though vividly, felt, but ideas are clearly known as the foundation of all the sciences); that these ideas are indeed efficacious because they act in the mind, they enlighten it and make it happy or unhappy, which is evident by the pain that the idea of the hand causes in those who have had an arm cut off. (OCM II 103, LO 322–3) The argument re-appears in a somewhat different guise in Éclaircissement X, where Malebranche is arguing for the existence of a universal reason shared by all minds: I am certain that the ideas of things are immutable, and that eternal laws and truths are necessary – it is impossible that they should not be as they are. Now, I see nothing in me of a necessary or immutable nature – I am able not to be, or not to be such as I am; there might be minds unlike me, yet I am certain that there can be no mind that sees truths and laws different from those I see – for every mind necessarily sees that twice two is four, and that one’s friend is to be valued more than one’s dog. It must be concluded, then, that the reason consulted by all minds is an immutable and necessary Reason. (OCM III 130, LO 613–14) In the first of the Entretiens, Ariste has difficulty grasping the reality of the realm of ideas, and suggests that the contents of the intelligible world simply lapse back into nothingness when one is not actually thinking of them. If this were true, Theodore retorts, ideas would indeed be ‘peu de chose’: True, Aristes. If you gave being to your ideas, if an idea requires but the blink of an eye to annihilate it, it is indeed a slight thing. However, if they are eternal, immutable, necessary, in a word, divine – I mean the intelligible extension from which they are formed – surely they will be more considerable than that matter which is inefficacious and absolutely invisible by itself. (OCM XII 40, JS 12) Our perceptions are modes of our souls, and are contingent, fleeting, changeable things, readily altered or even annihilated. Ideas are eternal, immutable, and necessarily existent. They are also, in an important sense, ‘infinite’,33 whereas modes of our souls are finite particulars. If relations
59
THE VISION IN GOD
of ideas like 2 + 2 = 4 are necessary and eternal truths, the constituent ideas themselves must be eternal and necessary. But if so, they cannot be identified with our perceptions. When challenged by Arnauld on this point, Malebranche invokes the Christian Platonism of Augustine.34 According to Augustine, we are told, ideas are eternal, immutable, necessarily existent and common to all spirits – i.e. wholly distinct from the modes of human souls. Ideas, for Augustine, are: … certain stable and immutable first models or archetypes of all things, which have not been created, and which in consequence are eternal and remain always the same in the eternal Wisdom which contains them. (OCM IX 915) The preface of the 1696 edition of the Entretiens involves another sustained attempt to invoke the authority of Augustine.35 Since the Vision in God appears paradoxical to many readers, writes Malebranche, I thought it best to remind readers of Augustine’s notion of ideas as the eternal and immutable archetypes of all creatures (OCM XII 10–11). God needs no model external to Himself to create the universe, and knows the universe He has created simply by His immediate familiarity with His own perfections, in which creatures participate. For Augustine, then, ideas (archetypes in the mind of God) are quite distinct from perceptions (modes of our souls). I admit, says Malebranche, that Augustine’s version of the Vision in God only applies to our knowledge of eternal and necessary truths (arithmetic, geometry, morality), but if he had been aware of seventeenth-century advances in natural philosophy he would have extended his theory and arrived at my conclusions (OCM XII 17–18). If ideas are divine archetypes, then by having access to ideas we can gain knowledge of things. Certain types of sceptical objection can be laid to rest once and for all. It is, after all, logically impossible that a creature should fail to conform to God’s idea of it, which is the model or archetype after which it was (and is) created.36 On this Platonist theory, an idea is not a copy of a thing, capable of greater or lesser fidelity to the original; on the contrary, the thing is a copy of its idea, and cannot (because of continuous creation) depart from the original. This epistemological advantage of the Vision in God would be emphasised by Malebranche in his polemical exchanges with Arnauld. The contrast Malebranche emphasises is between ideas on the one hand and all the modes of the soul on the other. The most obvious and
60
THE VISION IN GOD
striking modes of the soul are of course sensations, but there are also pure perceptions, inclinations and habits. Pure perceptions are the modifications our souls undergo when we become aware of ideas. Malebranche notes their existence, as he must, but has little to say about them because, as he says, they don’t deeply affect the soul, and have no characteristic phenomenology.37 When I am deep in thought about the properties of the circle or the requirements of justice it is the idea that occupies my consciousness; my own act of awareness is as it were transparent. But all modes of the soul (sensations, pure perceptions, inclinations, and habits) are contingent (as modes of a contingent substance, they could hardly be otherwise), changeable, and capable of ceasing to be. As such they all fall equally within the scope of Malebranche’s argument. The contrast between ideas and sensations is therefore, for Malebranche, no more than a corollary of his general distinction between ideas on the one hand and all modes of the soul on the other. The idea-sensation antithesis is, nevertheless, central to Malebranche’s account of the soul and self-knowledge. On the Cartesian theory, our soul contains ideas of sizes, shapes and motions, which are ‘clear and distinct’, and ideas of colours, sounds, tastes, smells, etc, which are confused. It is natural, on such a theory, to think of all these ideas as representations, and to think of their clarity and distinctness as admitting degrees. A modern reader might think of photographs taken in better or worse light, from close up or far away, in good or bad focus. Malebranche will have none of this. He insists on a sharp contrast between sensations, which are modes of our souls and represent nothing beyond themselves, and ideas, which are ‘in’ God and represent objects.38 His rejection of the monde intelligible hypothesis H4 thus carries with it an absolute denial of the thesis that intentionality is the mark of the mental.39 All I can tell when I have a sensation such as hot or cold, red or blue is that my soul is capable of being modified in such a way. When I think of a triangle, a square or a circle, by contrast, I gain knowledge of the real properties of material creatures, actual or possible. But I gain this knowledge by way of the idea itself (the archetype in the divine intellect), not by way of the mode of my soul which is my pure perception of that idea.
Intelligible extension Malebranche’s theory is clearly a Christian version of Platonism, identifying ideas with archetypes in the divine intellect. This would suggest that God’s ideas are universals: that His mind contains the ideas or forms of Cat, Dog,
61
THE VISION IN GOD
Horse and Cow, but not ideas of each individual cat or dog, horse or cow. But the theory of creation pulls in the opposite direction. It is individual cats and dogs, horses and cows that God created in the beginning and continues to sustain in being. Since He must know what He is doing, He must have ideas of each and every individual creature. This objection was urged by Arnauld,40 and has been endorsed by Gueroult.41 But if God has ideas of each and every individual creature, and it is these ideas which are the immediate objects of our experience, all manner of bizarre and counterintuitive conclusions follow. Since, on this view, every creature has its idea, the intelligible world becomes as complex, confused and changeable as the material world. The moon illusion provides a good illustration of this point. As I watch the moon sail across the heavens, I see it as larger at the horizon than when directly overhead. The direct or immediate object of my experience (the idea) is the bearer of these phenomenal properties. So the idea, we would have to infer, is itself changeable. But astronomy teaches us that the moon does not in fact change in size. So if God has an idea of the moon it seems that it must both stay the same size (to serve as the model or archetype of an unchanging object) and vary in size (to serve as the direct object of our experience). A contradiction threatens to open up in the Malebranchian theory of ideas. Similar objections can be raised for any number of cases of perceptual illusion and perceptual relativity. This objection is raised and addressed in Éclaircissement X of the Recherche (OCM III 151–5, LO 626–8), which marks a significant modification in Malebranche’s account of ideas. Although nothing ‘in’ God is actually figured or mobile, He must be able to understand figured and mobile things ‘in’ Himself: But to clarify this matter, it must be realized that God contains in Himself an ideal or intelligible infinite extension; for since He has created it, God knows extension, and He can know it only in Himself. (OCM III 151–2, LO 626) For the entire material creation, a single idea or archetype, étendue intelligible, is sufficient. Since God created the entire material universe ex nihilo, and understood what He was doing, He must have had a model or archetype in His mind, and this is intelligible extension. The point is repeated in the controversy with Arnauld (OCM VI 117) and in the second Entretien (OCM XII 51, JS 21). Given this single idea of intelligible extension, God sees all possible material creatures as modes
62
THE VISION IN GOD
or modifications of extension. That is, He sees how extension must be modified to produce sun and moon, trees and flowers, cats and dogs. The infinite intellect of God thus contains all possible material creatures, and all possible material universes, simply by having the idea of intelligible extension. The actualisation of some subset of all these possible creatures depends of course on the will of God, which brings about an optimal expression of the divine attributes.42 Malebranche goes on, in Éclaircissement X, to explain what happens when we see particular bodies, and how we can observe change and motion when the intelligible realm is eternal and immutable. We see a given body, he explains, … when its idea, i.e. when some figure composed of intelligible and general extension, becomes sensible and particular through color or some other sensible perception by which its idea affects the soul and that the soul ascribes to it, for the soul almost always projects its sensation on an idea that strikes it in lively fashion. Therefore, there need be in God no sensible bodies or real figures in intelligible extension in order for us to see them in God or in order for God to see them in Himself. It is enough that His substance, insofar as it can be participated in by the corporeal creature, should be able to be perceived in different ways. (OCM III 152, LO 626) For me to see the full moon, all that is required is to perceive a circular portion of intelligible extension silver-ly, i.e. for that sensation of colour (actually a mode of the soul) to be ‘projected’ onto a suitable part of étendue intelligible. When the moon is at the horizon human viewers will perceive a greater portion of intelligible extension silver-ly than when it is overhead. A similar account will handle our perception of motion: Likewise, if, as it were, a figure of intelligible extension made sensible by color should be taken successively from different parts of this intelligible extension, or if a figure of intelligible extension could be perceived as turning on its center or as gradually approaching another, we would perceive motion in an intelligible or sensible figure without there being any actual motion in intelligible extension. OCM III 152, LO 627)
63
THE VISION IN GOD
Intelligible extension is like a blank canvas or cinema screen onto which different sensations are projected. For me to see a cricket ball is just to see a circular portion of intelligible extension red-ly. To see the cricket ball flying across my visual field is just for me to see successive circular portions of intelligible extension red-ly. Motion can thus be represented ‘in’ intelligible extension without literally being present there. Since all the parts of intelligible extension are perfectly homogeneous, ‘they may all represent any body whatsoever’. Once we understand this, says Malebranche, we will see that: It should not be imagined that the intelligible world is related to the sensible, material world in such a way that there is an intelligible sun, for example, or an intelligible horse or tree intended to represent to us the sun or a horse or a tree, or that everyone who sees the sun necessarily sees this hypothetical intelligible sun. (OCM III 153, LO 627) Arnauld thought that Malebranche had simply abandoned his earlier theory of particular ideas and replaced it with the completely different and incompatible theory of intelligible extension.43 This accusation is echoed by some of the commentators.44 Malebranche replies that the theory of intelligible extension is a clarification of his earlier position, not a rejection of it. This opinion too has found support among the commentators.45 What matters, however, is not Malebranche’s alleged change of mind but how the new theory is meant to work. Intelligible extension, although it doesn’t actually contain within itself all material creatures, contains them virtually, in the sense that it contains the means of engendering or constructing them. 46 But, as a number of the commentators note, particular ideas don’t drop out of the story.47 Malebranche continues to speak, for example, of the ideal arm of the amputee (e.g. at OCM XIX 910), or of my ideal hand affecting my soul with ideas of both warmth and colour (OCM XVII–1 287–8). So Malebranche’s theory involves both étendue intelligible (the archetype or blueprint of all possible bodies) and ideas of the particular bodies which are the objects of our experience. Is this consistent? Malebranche, it should be clear, needs both the single ‘infinite’ idea of étendue intelligible, and the many ‘finite’ ideas of particular created things. The former is timelessly present in the divine intellect; the latter exist – with appropriate temporal qualifiers – in the divine will. Since the divine will is not blind, God must, as Arnauld insisted, have ideas of particulars.
64
THE VISION IN GOD
The idea of étendue intelligible is a precondition of all ideas of bodies, but it could never serve as a substitute for them. When we do geometry, we consult intelligible extension alone, conceiving of triangles, squares and circles merely as possible modes of extension.48 In our experience of bodies, by contrast, intelligible extension is necessary but not sufficient: we also need ideas of particulars, impressed on our souls through a sort of natural revelation of the divine will. What we see when we see a body is intelligible extension, ‘painted’ red or green or blue by the projection of a mode of our own souls. As Theodore explains in the third of the Conversations Chrétiennes, ‘It is evident that to see a body is nothing else but to be modified with different sensations of colour by diverse parts of intelligible extension’ (OCM IV 75). As Malebranche explains to Régis, we visually distinguish and differentiate bodies by their colours. To make such judgements requires that ‘the idea of extension, according to its different parts, modifies my soul, here with one colour and there with another’ (OCM XVII–1 281). It would be easy to cite any number of passages in which Malebranche speaks of the parts of intelligible extension, and uses the notion to explain, for example, our capacity to distinguish and differentiate objects, our susceptibility to perceptual illusions, and our perception of motion. The most natural reading of such passages is to think of étendue intelligible as a blank canvas, or, if you prefer, a featureless block of marble awaiting the sculptor’s chisel.49 But Malebranche firmly and explicitly denies that intelligible extension itself consists of spatial parts, i.e. that it is itself literally extended. In the first of the Three Letters to Arnauld he insists that the ‘intelligible places’ where the soul sees colours are not real places. The greater parts do not occupy more space than the smaller, because they do not occupy space at all (OCM VI 211–12). In the fourth of the Entretiens Theodore takes some pains to distinguish material spaces from intelligible spaces. The spaces we see, he maintains, ‘are simply intelligible spaces which fill no place (OCM XII 95, JS 58).’ ‘I never’, Malebranche writes in response to Arnauld, ‘thought that the idea of length, breadth and depth was itself long and broad and deep’ (OCM VI 242).50 We appear by now to have arrived at a hopeless tangle. Malebranche frequently employs the blank canvas analogy, and speaks freely of parts or portions of intelligible extension. This is not a mere lapse or aberration on his part, but an essential aspect of his explanatory theory, his attempt to extend the Vision in God to our ordinary sensory perceptions of bodies. But when challenged by Arnauld he insists that he never dreamed 65
THE VISION IN GOD
that intelligible extension itself consists of spatial parts. Intelligible extension is the idea of matter, the archetype God consults in His ongoing creation of matter. It must therefore contain, for Malebranche, all the perfections of matter, albeit in an ‘eminent’ or higher form. Its parts, he insists in reply to his critics, are intelligible parts and not spatial parts.51 The problem is, of course, that at the conceptual level the idea of extension does not consist of parts at all. There is not the least reason to suspect that the concept of extension is subject to further analysis, and every reason to regard it as a primitive. Malebranche, it appears, has been guilty of an illegitimate slide from ideatum to idea. It belongs to the concept of extension that any instance of extension will consist of parts. It does not follow that the concept consists of ‘intelligible’ parts, or that these stand in some sort of isomorphism with the spatial parts of the body.52 Malebranche’s critics sense that he is floundering at this point, and move in like sharks for the kill. Foucher thinks that resemblance is a necessary condition for representation, and hence that the theory of ideas requires something like the blank canvas theory, with spatial relations represented by spatial relations.53 Arnauld thinks that Malebranche has identified intelligible extension with God’s immensity, and thus made God literally extended.54 Locke finds it inconceivable that something spiritual (and hence unextended) should represent an extended thing55 . And Malebranche’s last critic, De Mairan, suggests that Malebranche’s theory slides into Spinozism (OCM XIX 861). If étendue intelligible is infinite and eternal and necessarily existent, De Mairan asks, and is the archetype of matter, why isn’t matter infinite and eternal and necessary? There is no way to save Malebranche’s theory as it stands. The reason is simple. He requires intelligible extension to play two distinct and incompatible roles. On the one hand, it is the logical concept of extension, which is not the sort of thing that could intelligibly be said to be extended. If it consisted of parts at all, they would be conceptual parts (as ‘animal’ is a part of ‘bear’ and ‘male’ is a part of ‘bachelor’), but there is no reason to suppose it consists of parts in this sense. Possession of this concept is a necessary but not a sufficient condition for certain types of perceptual experience. On the other hand, intelligible extension plays a role akin to the Kantian notion of space as the pure form of outer intuition, i.e. as an a priori particular and not a concept at all. It is particulars, Kant reminded us, that divide into parts, and Malebranche’s blank canvas analogy positively lends itself to interpretation as an a priori particular. Lacking Kant’s doctrine of space, and his associated distinction between pure and schematised concepts, Malebranche could not help but fall into confusion.56 66
THE VISION IN GOD
Efficacious ideas Malebranche’s attempts to explain how the divine archetypes can act on our minds lead, according to Martial Gueroult, not to a clear solution but to a ‘labyrinth’.57 How, for example, does étendue intelligible produce our perceptions of bodies? We can represent the action as that of the idea as such, as distinct from the substance of God and of the other archetypes. On this account, the effect is conceivable, but not the manner of the causal action. Or we can say that God Himself is the cause. Here the causal action is intelligible, but not the effect – it is after all intelligible extension that we see, not the substance of God Himself. Malebranche wants to maintain both that God and God alone acts on my soul when I perceive, but also that the Vision in God is not a Vision of God. Can these two theses be reconciled? According to Robinet, Malebranche came increasingly to stress the first of Gueroult’s two models at the expense of the second. This theory of efficacious ideas, we are told, is Malebranche’s preferred model from 1695.58 Ideas are characterised not just as eternal, immutable and necessarily existent, but as efficacious, i.e. as possessing precisely the power of acting on our souls that is denied to bodies. To see a body, Malebranche writes in his Réponse à M. Régis, written in 1693 and included in the 1700 and 1712 editions of the Recherche, is: … to have present to the mind the idea of extension which touches it or modifies it with different colours: for one does not see them [bodies] directly or immediately in themselves. (OCM XVII–1 282) When I speak of extension, he continues, I mean intelligible extension, the archetype of matter: ‘For it is clear that material extension cannot act efficaciously and directly in our minds. It is absolutely invisible of itself’ (OCM XVII–1 283). In a letter to De Mairan from the last year of Malebranche’s life, the causal agency of ideas is still more clearly highlighted: It is the idea of the hand that affects with pain the soul of the amputee: the ideatum, that is to say the hand, no longer exists, it has been eaten by worms. It is the idea of a spectre which frightens a madman, its ideatum does not exist at all. (OCM XIX 910)
67
THE VISION IN GOD
But the theory of efficacious ideas only raises further questions. How, Jolley asks, can an abstract object like an idea have causal powers? 59 And what, Nadler asks, is the ontological status of these ideas? 60 Are they substances or modes? How do they relate to the substance of God? In his refutation of the ‘production’ theory, Malebranche was wary about specifying the precise ontological status of ideas. They are, he insists, ‘real beings’, and real beings of a ‘spiritual’ nature, superior to bodies and beyond our power to create (OCM I 423, LO 222). If my critic objects that an idea is not a substance, he writes, I would agree – but it is still a ‘spiritual thing’, and a spiritual thing can neither be created by the human soul ex nihilo nor made from a material thing (an impression on the sense organs). In the first of the Three Letters to Arnauld, the status of intelligible extension is explicitly addressed. Notwithstanding the axioms of the philosophers, we are told, it is neither a mode nor a substance (OCM VI 245). This is a puzzling remark. Other philosophers of the seventeenth century (e.g. Gassendi and Locke) challenged the metaphysics of substance and mode, but Malebranche is generally faithful in his allegiance to it.61 So why does he appear to depart from it in this case? Ideas, Malebranche thinks, cannot be modes of God because God’s substance is not susceptible to modification (OCM III 149, LO 625).62 And we cannot simply identify ideas with the substance of God, for a number of reasons. To see an idea is not to see God. And to see one idea (e.g. étendue intelligible) is not to see another (e.g. the archetype of the human soul). But if A is identical to B and B is identical to C, A is identical to C. But if Malebranche has reasons not to identify ideas with the substance of God, he also has pressing reasons forcing him to make that identification. Ideas act on the soul. But only God can act on the soul. So ideas cannot be really distinct from God.63 Malebranche’s final attempt to extricate himself from this apparent contradiction is to say that ideas can be identified with the substance of God, but with an all-important qualification. God’s ideas, he tells us in the preface to the Entretiens, are ‘nothing more than the essence of God, insofar as creatures can imitate it or participate in it’ (OCM XII 12).64 Intelligible extension, for example, is ‘the substance of God as representative of body’ (OCM XII 184, JS 136–7). Every creature must participate, more or less, in one or more of the divine perfections – it is not just man who was created in the image of God (OCM IV 64). In creating the human soul, as in the creation of matter, God needs only to look to His own perfections to find intelligible archetypes. The point is 68
THE VISION IN GOD
expressed most clearly – with a rare reference to Aquinas – in Chapter Eleven of Book Four of the Recherche: God knows the nature of the soul clearly because He finds in Himself a clear and representative idea of it. God, as Saint Thomas says, knows His substance or His essence perfectly, and as a result He discovers all the ways in which created things can participate in His substance. Hence His substance is truly representative of the soul, because it contains its eternal model or archetype. For God can only draw His knowledge from Himself. (OCM II 97–8, LO 319) Arnauld thinks that Malebranche’s theory entails the impossible conclusion that we do indeed see God, i.e. that the Vision in God collapses into a Vision of God.65 If what we perceive is an intelligible world of ideas, and these ideas are not really distinct from God, then surely we are seeing God? Malebranche had already done his best to address this problem in his replies to objections in Éclaircissement X of the Recherche. The fourth objection quotes Saint John to the effect that no one has seen God. Malebranche replies that: I answer that seeing His creatures in Him is not really seeing God. Seeing the essences of creatures in His substance is not seeing His essence, just as merely seeing the objects it represents is not seeing a mirror. Seeing the essence of God, not in its absolute being, but in relation to creatures or insofar as it is representative of them, is not seeing the essence of God. (OCM III 155, LO 628) So ideas are God, but only insofar as His essence is participable or imperfectly imitable by creatures (OCM III 149, LO 625). The point is repeated in the Réponse à M. Régis (OCM XVII–1 293). If forced to say whether ideas are substances or modes, we must say they belong to the category of substance, since God is a substance and they are not really distinct from God.66 He could not exist without His perfections, nor without perfect knowledge (a) of those perfections and (b) of the capacity of creatures to participate in them. So He could not exist without the ideas of creatures. But the ideas could not exist without Him. So there is no real distinction between God and his ideas. Gueroult’s two models, contrasting God’s action with efficacious ideas, are just two different
69
THE VISION IN GOD
ways of saying the same thing. In the final analysis, he has hit on a distinction without a difference. How much sense can we make of this doctrine? The continual references to participation sound Platonic, and bring back thoughts of Plato’s forms. But this in turn only revives age-old worries about selfpredication. The form of F, Plato thought, was ‘the F itself’; other things are more or less F by participation in the form. But Malebranche explicitly denies self-predication – at least in the case of intelligible extension.67 The form or idea of extension is not itself extended, not something that could intelligibly be said to be long or broad or deep. So how can it be the form, model, or archetype of extension? How can God, as it were, look to the model and say to Himself ‘I’ll create something like that’? What can ‘like’ mean in such a context? Malebranche says that the ideas of things ‘resemble them in some manner’, but then denies the obvious interpretation of resemblance.68 He will of course say that extension is present in God not formally but eminently, i.e. in a higher form. He can insist on the orthodoxy of his position, and quote Aquinas. God is Himself unextended, but He creates extended things. But God must know (i.e. understand) what He is doing in creation. So there must be something in God that represents extension to Him. The premises of this argument are impeccably orthodox; the inference seems valid. But the conclusion is only a thin, almost skeletal existential claim, of no explanatory power. We seem to have arrived at the limits of our comprehension. In the case of intelligible extension, we might begin to develop – albeit speculatively – a reply along the following lines. In the infinite intellect of God there are no figures, only equations. While human minds naturally turn to geometry, God prefers algebra and analysis. In creating circular bodies, He looks to the formula a² + b² = r², not to any literally extended model. But the mathematical formula does ‘resemble’ the geometrical figure in the crucial sense that it contains what is intelligible about the latter. So Malebranche can defend what I called an Epistemological Intelligibility Principle (EIP) while rejecting Watson’s Epistemological Likeness Principle (ELP). Unfortunately, this only gets us so far. In creating the bodies of the solar system, God looks to the formula of the sphere and not to a three-dimensional model or set of blueprints. But He must still say to Himself, ‘The system will consist of nine of these (approximate) spheres in elliptical orbits about a larger central sphere’. The formula for the sphere can be present in the divine intellect without extension, but could the plans of the whole solar system be just a more complex set of mathematical equations? Even if this were possible, such a vast and complex formula would provide only the 70
THE VISION IN GOD
blueprint for a possible material creation. The actual creation of concrete particulars is a matter for God’s will rather than His intellect, and this will – if it is not to be blind – seems to require ideas of particulars.
Nadler’s Malebranche In Malebranche and Ideas, Steven Nadler has challenged the traditional view of Malebranche as an indirect realist, an advocate of a representative theory of perception.69 Malebranche’s ideas, Nadler claims, are not the sense data of the empiricists; rather, they are logical concepts. To have an idea of X is not to have a little picture of X before the mind’s eye; it is to comprehend the essential properties of X, and thus to see what properties X must possess and what properties X cannot possess. The negative part of this case is clearly correct. Whatever Malebranche’s ideas are, they cannot be sense data. Any merely sensible image present to the mind’s eye would just be a mental particular, not a source of insight into necessary and universal truths. But Nadler’s positive claim – that ideas are logical concepts, or sets of necessary and sufficient conditions – involves, to my mind, just too much special pleading and selective reading of the texts.70 Could all the commentators, from Arnauld and Locke down to Nadler himself in his earlier book, have been so mistaken for so long? And how is Nadler going to explain away all those passages71 in which Malebranche does represent ideas as direct or immediate objects of quasivisual perception? There are just too many places in which Malebranche speaks of ideal arms and legs, an ideal sun and moon – i.e. ideal particulars – as immediate objects of perception. On Nadler’s reading, Malebranche’s theory of ideas is primarily an account of pure intellection, and the relation between idea, mind and object is a cognitive one – to have an idea of X is to know the properties of X.72 For pure intellection, the Nadler interpretation seems to work well. But to apply his interpretation to Malebranche’s account of sense perception, Nadler finds himself having to torture the text. Malebranche says, for example, that we do not see bodies ‘immediately’ and ‘in themseves’; rather, we see them by the mediation of ideas. What this means, says Nadler, is that sense perception involves judgement, and judgement requires concepts, i.e. that we couldn’t see shapes unless we had the concept of extension and of its possible modifications. Nadler’s Malebranche is thus a sort of precursor of Kant, reminding his readers that perceptions without concepts are blind. Now I have no quarrels with the truth of all this. Perception does involve judgement; judgement does require concepts. Nor is it a thought
71
THE VISION IN GOD
that would have been alien to a Cartesian – indeed, it is precisely what Descartes was trying to show in the famous example of the wax in the second Meditation. But this insistence on the central role of concepts in sense perception cannot be all that Malebranche had in mind. In the first place, Malebranche’s denial that we see bodies ‘immediately’ and ‘in themselves’ is clearly meant to be contrastive. Bodies are not, but ideas are, the objects of such immediate awareness. And logical concepts just aren’t the right sort of things to serve this role. If we are to have an account of perception that can do justice to the phenomena, there is no getting away from an ideal realm consisting of sensible particulars. To extend the theory of ideas from the intellect to the senses Malebranche needs some Kantian apparatus that he simply does not possess. He needs space as a ‘form of intuition’, an a priori particular that is simply ‘given’ to the perceiving mind rather than grasped through concepts. And he needs the Kantian notions of schemata and schematized concepts. On this view, seeing a triangle in the mind’s eye is not just a matter of possessing a concept, but of exhibiting it to intuition by a process of construction.73 The logical concept still serves a role (e.g. guaranteeing universality) but it is schematised concepts that I will be employing all the time in my perceptual judgements. I don’t claim, of course, that Malebranche had all this Kantian apparatus (although there are important constructivist strands in seventeenth-century mathematics), merely that I don’t see how his theory can work without it. To account for the role of the understanding in visual perception, we will need, it seems, a threefold distinction between ideas(l) or logical concepts, ideas(s) or schematised concepts, and ideas(v) or visible particulars, which are schematised concepts made manifest to sight by means of colour. Without ideas(s) there could be no ideas(v); I cannot see the sun as a yellow circle in the sky without the capacity to imagine a circle. Without ideas(l) there can be no ideas(s); I cannot mark out a circle in my mind’s eye unless I possess the pure concept of extension, étendue intelligible. Malebranche, of course, nowhere makes these distinctions, but his theory cannot work without them. In conclusion, we can learn much from Nadler, and endorse many of his insights, without accepting his central contention that Malebranche is a direct realist about sense perception. He is right to deny that Malebranche’s ideas are like the sense data of the empiricists. If Malebranche comes close to any form of idealism, it is Kantian rather than Berkeleyan. He is right to emphasise that Malebranche’s theory of ideas is primarily an account of the intellect, and only secondarily applied to sense perception. This puts Malebranche squarely in a tradition 72
THE VISION IN GOD
stemming from Plato via Augustine. And he is right to stress the role of the understanding, and hence of concepts, in the processes of discrimination, recognition and anticipation that are integral to sense perception. But Nadler cannot be right to deny that, for Malebranche, the direct or immediate object of perception is an intelligible world of intelligible particulars, the existence of which is logically independent of the material world of material particulars.74 The texts simply will not bear such an interpretation.
73
THE DISPUTE WITH ARNAULD
4 THE DISPUTE WITH ARNAULD OVER THE N AT U R E O F I D E A S Malebranche’s early relations with the Port Royal group were amicable, with Arnauld in particular responding favourably to the Recherche on its first appearance in 1674.1 Arnauld clearly hoped to draw the promising young philosopher into a Cartesian–Augustinian–Jansenist alliance against the Aristotelian–Thomist–Jesuit traditions of the schools.2 In the light of Malebranche’s later works, this might seem to have been a forlorn hope. But the views of the early Malebranche, as Robinet has shown,3 were close to Jansenism in a number of important respects. At this stage, Malebranche still accepted the omnipotent but inscrutable ‘Dieu caché ’ of the Jansenists, and was correspondingly wary about the incursion of philosophical rationalism into theology. The publication in 1678 of the Éclaircissements to the Recherche marked a radical change in the situation, and precipitated the final and definitive rupture between Malebranche and the Jansenists. In Éclaircissement VIII Malebranche launched an outright attack on the voluntarism of the Jansenists, arguing that such a position ‘overthrows everything’, science, religion and morality alike (OCM III 84, LO 586). A God worthy of the name must rule by Wisdom, not by mere omnipotent Will. His volitions must be in accordance with an Order that is in principle accessible to all minds.4 It follows, says Malebranche in Éclaircissement XV, that God’s actions in the domain of grace, as in that of nature, must be in accordance with volontés générales rather than volontés particulières. There are simple and universal laws of grace as there are laws of nature (OCM III 221, LO 666–7). It was this claim that scandalised Arnauld, and led – by way of an acrimonious meeting between the two men in 1679 – to the publication in 1680 of Malebranche’s Traité de la Nature et de la Grâce. For Arnauld, Christianity requires us to believe in a God who cares for particular human beings, and who distributes grace to individual men and women by means of volontés particulières. We say, of course, that God is wise 74
THE DISPUTE WITH ARNAULD
and just, but we should not presume that there are independent and universally binding norms of conduct binding God’s will as well as ours. Whatever God wills is wise and just precisely because He wills it. So why, given that his primary disagreement with Malebranche was over the theological topic of grace, did Arnauld choose to begin his attack in 1683 with the Vraies et Fausses Idées? At first sight, the metaphysical issue of the nature of ideas seems far removed from the theological controversy about grace. Malebranche, indeed, accused Arnauld of perversely changing the subject in choosing to attack the most abstract and difficult part of his entire system, rather than engaging directly with the Traité (OCM VI, 18). In fact, however, Arnauld is not merely being perverse or wilful – the two subjects are more closely related than one might at first suppose. But we must return to this subject later.
Arnauld’s Vraies et Fausses Idées Arnauld’s objective in writing Vraies et Fausses Idées (VFI) can be simply stated. We shall demonstrate, he claims, that … what the author of the book THE SEARCH FOR TRUTH says about the subject [ideas] is based on nothing more than false prejudices, and that nothing is more groundless than his claim that WE SEE ALL THINGS IN GOD.5 Arnauld sets out to refute the Vision in God by discrediting Malebranche’s conception of ideas as êtres représentatifs. If there are such ideas, with properties of their own, existing independently of our perceptions, then questions can be asked regarding the ontological status of these ideas, and the Vision in God will come into play as one hypothesis among others. But if our ideas simply are our perceptions, i.e. modes of human souls, their humble ontological status is already settled, and the Vision in God can be dismissed as an unnecessary answer to an illconceived question. The fundamental assumptions of VFI can be stated very briefly. Most basic of all are the metaphysics and epistemology of Descartes: Arnauld is at some pains to present himself as the true Cartesian and Malebranche as the dangerous innovator.6 Next comes the appeal to the evidence of introspection: Arnauld shares with Descartes the principle that the mind has certain knowledge through introspection of all of its states and operations.7 Finally there is a version of Ockham’s razor: God will not
75
THE DISPUTE WITH ARNAULD
perform by complex and roundabout means what He can perform by simple and direct ones.8 Since Malebranche explicitly makes use of the same principles, Arnauld feels himself entitled to assume Malebranche’s assent to all these premises. What, asks Arnauld, are our perceptions for ? The answer is obvious: their primary function is to enable us to know the existence and properties of the bodies surrounding us. But if this can be achieved by simple means, he asks, what need is there of complex ones? The simplest means of all is for God to give each soul the power or faculty of forming ideas on the occasion of certain physical events in its associated brain. These ideas are our perceptions of external objects. There is therefore no need of Malebranche’s ideas, representative entities distinct from perceptions: God did not will to create our soul and to put it into a body surrounded by countless other bodies, without also willing that it be capable of knowing bodies, and consequently without also willing that bodies be conceived by our soul.9 But all of God’s volitions are efficacious, so He has created our souls with just this capacity. Now God does not accomplish by complex and roundabout means what can be brought about by simple and direct means. So there is no need to posit an intermediate realm of ideas; it is simpler to suppose that the mind knows bodies directly than that it only knows them indirectly by means of representations. Although my mental acts are modifications or modes of my soul, they are also essentially representative of objects. The soul, says Arnauld, is by its nature a thinking substance. The soul thinks because that it is essence: there is nothing more to say. But thought is always thought of something or other: So, since there can be no thought or knowledge without an object known, I can no more ask what is the reason why I think of something than why I think, since it is impossible to think without thinking of something. But I can very well ask why I think of one thing rather than another.10 In other words, intentionality is simply built into our mental states: a particular modification of the soul is essentially a perception-of-F. Such an account of the mind, Arnauld contends, has two great virtues to recommend it. It is the simplest account that does justice to the phenomena. And it is in conformity with the authentic testimony of
76
THE DISPUTE WITH ARNAULD
introspection, which tells me that my perceptions are perceptions of things, not perceptions of êtres représentatifs distinct from and independent of my mind. If, says Arnauld, Malebranche had reflected on the contents of his own mind when perceiving a cube, ‘I am sure that he could not have seen there anything other than the perception of the cube, or the cube objectively present to the mind’.11 Arnauld thus has an ‘act’ theory of ideas and a direct realist account of perception. We must still, however, take account of one small quirk of terminology that has misled some of the commentators. Arnauld does speak in VFI of the idea as the immediate object of perception and of the external body as (normally) the mediate object.12 Although he takes some pains to explain what he means by this, the damage had been done, and some critics inferred (erroneously) that he meant by those terms more or less what others had meant by them. What Arnauld means is just that every mental act involves a ‘virtual reflection’ on itself, as well as being directed onto its proper intentional object. The mental act of perceiving the sun is not itself the object perceived, although of course it may become the object of a distinct act of introspective reflection.13 So although it is trivially true that I can’t see the sun without having an idea of the sun (because the idea is the seeing), it doesn’t follow that the idea is the only thing that is directly or immediately perceived in the usual sense of those words. Although Arnauld identifies ideas with perceptions, he admits a sort of dual aspect theory. The same things can be conceived and spoken of in two different ways. In some contexts it will be appropriate to speak of perceptions (when the emphasis is on their ontological status as modes of the soul); in other contexts, we should speak of ideas (when the emphasis is on their representative role). This is made perfectly explicit in VFI: I have said that I take the perception and the idea to be the same thing. Nevertheless it must be noted that this thing, although only one, has two relations: one to the soul which it modifies, the other to the thing perceived in so far as it is objectively in the soul; and that the word perception indicates more directly the first relation and the word idea the second.14 Given the superiority of Arnauld’s direct realism on the two crucial counts of (a) simplicity and (b) conformity with the testimony of introspection, we must ask why Malebranche felt obliged to introduce his representative ideas as intermediates between the mind and its objects.
77
THE DISPUTE WITH ARNAULD
Arnauld suggests that Malebranche was led astray by a misleading analogy with physical optics. He is assuming, Arnauld suggests, that the object of perception must be present to the mind in some literal sense, if not itself then by some likeness or proxy. But this is a mere prejudice. All the optical talk about images has led Malebranche (and others) astray.15 In the first place, when I look at my face in a mirror, it is my face that I see, not some distinct entity called an image. Even if one admits that an image in some literal sense, i.e. a miniature picture of the object, plays an essential role in the causal story of how vision occurs, this story (concerning the production of retinal images) lies wholly on the physical sense of the mind–body divide. There can be no resemblance across that divide. To attempt to explain the mind’s perceptions in terms of little pictures is to put the cart before the horse, to reverse the proper order of explanation. A picture does not wear its meaning on its sleeve: the intentionality of pictures (and for that matter of spoken and written language) is a secondary affair, dependent on the primary and original intentionality of the mind’s own perceptions. Unless our mental states had this original and intrinsic intentionality, no picture, and no written or spoken word, would mean anything at all. Malebranche’s ideal theory, Arnauld contends, is completely selfdefeating. He has set out to explain how our minds perceive external objects. In other words, he has started with the common sense presumption that we do perceive tables and chairs, trees and houses, and raised the question of how this can be the case. But his explanans is inconsistent with the original explanandum. The proper conclusion to derive from Malebranche’s theory is that we do not perceive trees and houses at all, but only variously coloured portions of the intelligible extension that is in God.16 One might expect Arnauld to press charges of scepticism against Malebranche at this point, but in fact scepticism about the external world plays a very small role in the whole debate. Presumably, Arnauld is aware that his own direct realism merely shifts the point of the sceptic’s attack. Granted, the sceptic can argue, that the direct object of a normal veridical act of perception is an independently existing body, how then do I know which if any of my perceptual states are thus veridical? If none of them are, then perhaps there is no external world at all, and I actually see no objects at all, all the objects I seem to see being merely intentional objects, having only ‘objective being’ in the Cartesian sense. In answer to such scepticism, Arnauld will merely refer his readers to the familiar argument of Descartes’ Meditation Six: if this were the case, God would be a deceiver, which is impossible.17 78
THE DISPUTE WITH ARNAULD
As for Malebranche’s use of the argument from illusion in support of his ideal theory, this is easy for Arnauld to meet.18 Given the scholastic notion of ‘objective being’, Arnauld can readily reply that one and the same object (e.g. the sun in Descartes’ famous example19), can exist objectively in different minds (or in the same mind at different times) with different properties, and that an object with no intrinsic or ‘formal’ reality at all (e.g. a green dragon) could nevertheless have objective reality, i.e. be an object of immediate perception. There need be no actual (i.e. non-intentional) bearer of merely apparent properties. The direct object of perceptual experience, Arnauld will say, is normally an independently existing external object, but on rare occasions there are perceptual states that have as their content representations of merely possible objects. For Arnauld, it is perfectly proper to say that the soul thinks because that is its nature. What this means is simply that it has this power or faculty because that is how God chose to create it. This is where explanation stops. To ask how perception is possible is like asking how matter can be divisible: Since it is the nature of the mind to perceive objects, some necessarily, so to speak, and others contingently, it is ridiculous to ask whence it arises that our mind perceives objects.20 Intentionality is indeed, as Brentano would later claim, ‘the mark of the mental’, a primitive and irreducible feature of mentality. For Arnauld, all mental states (including bodily sensations, perceptions of secondary qualities, and even pleasures) are intentional, i.e. are more or less clear and distinct perceptions. All attempts to explain intentionality – e.g. by positing impressions and ideas that are images or resemblances of objects – turn out to presuppose it. If one cannot perceive things, how will it help to have pictures of things? And if one can perceive things, what use are pictures of things?
Malebranche’s Réponse Malebranche’s reply to the Vraies et Fausses Idées is a great disappointment. It consists largely of complaints about Arnauld’s conduct, repetitions of points and arguments already made elsewhere, and accusations of misunderstanding – possibly wilful and malicious – on the part of his critic.21 The controversy between the two men was to drag on, its philosophical quality declining, until Arnauld’s eventual death in 1694.
79
THE DISPUTE WITH ARNAULD
Régis would enter the fray on Arnauld’s side in defence of what I have labelled the ‘mainstream’ Cartesian identification of ideas with perceptions. I shall not attempt to follow all the details; it suffices for our purposes here to extract the few points of philosophical substance made by Malebranche in his replies first to Arnauld and later to Régis. The state of the argument between Arnauld and myself, Malebranche begins, boils down to this. He believes that the modes of our souls are essentially representative; I believe that the modes of our souls are mere sensations, which represent to the soul nothing distinct from itself (OCM VI 50). We need no ideas (êtres représentatifs) to know the modes of our own souls, Malebranche explains – these are knowable in themselves, and evident to introspection. By contrast, our knowledge of external objects cannot be immediate in Arnauld’s sense. Arnauld thinks that it is ridiculous to enquire how our souls are able to perceive bodies. To those who find this incomprehensible, all Arnauld can reply is that it is the nature of mind to have this power or faculty. But, Malebranche retorts, the word ‘nature’ means no more in the mouth of a Cartesian than it did in the mouth of an Aristotelian (OCM VI 142). Talk about powers and faculties may be merely descriptive, a sort of innocuous place-holder for a real explanation, i.e. an intelligible account of how the substance in question comes to have the power or faculty. (Mechanistic accounts of gravity and magnetism would provide obvious analogies drawn from physics.) Alternatively, talk about powers and faculties may be meant as a sort of explanation in its own right; if so, such talk can only provide a pseudo-explanation that ultimately explains nothing at all. Arnauld’s claim that the mind thinks because it is its nature to think is manifestly a pseudo-explanation of this latter kind.22 What, Malebranche asks, is the source of Arnauld’s error? His problem, we are told, is that he has missed the significance of the deep metaphysical distinction between ideas and sensations (OCM VI 54). He should have heeded the distinction marked in the French language by the distinct terms connaître and sentir, between on the one hand knowledge of objects by way of their ideas and on the other hand mere immediate awareness of one’s own sensations. One can know ‘in oneself’ only such sensations as heat and cold, pleasure and pain, modes entirely devoid of representational content. Arnauld, says Malebranche, gives no explanation whatsoever of how a mere mode of my soul could be a representation of something quite different from itself, something it resembles in no way at all (OCM VI 91).23 As for all his apparatus of axioms and definitions, they are dismissed as merely question-begging. On Arnauld’s theory, every created spirit is trapped in the domain of its 80
THE DISPUTE WITH ARNAULD
own perceptions; all it really perceives are its own modes. Arnauld thus turns the soul into a monde intelligible in its own right,24 thereby violating the precept of Saint Augustine, ‘Do not say that you are your own light’ (OCM VI 63). But this theory, Malebranche insists, is not just impious and un-Christian; it is strictly unintelligible. How can the particular modes of a merely finite mind represent general truths, and infinity? The argument seems to have reached a stalemate – small wonder that no progress was being made. Malebranche is requiring, on a priori grounds, a very strict principle of intelligibility. For me to know something, it must in some way be assimilated to my mind, or represented by some proxy that is thus assimilated. Now bodies, mere chunks of extension, are not knowable in themselves, and are incapable of the right sort of union with the knowing mind. 25 Therefore there must be some intermediary that is (a) of such a nature that it is capable of entering into such a union with my mind, and (b) has some intelligible intrinsic intentionality – its very nature is to be an idea-of-F. Only the ideas in the divine mind meet these two joint requirements, so only the Vision in God can provide an intelligible account of the nature of mind. Arnauld would reply by simply denying the epistemological intelligibility principle (EIP) that is bearing the weight of Malebranche’s argument. For Arnauld, it is a peculiar but essential feature of the mind that its states can and do represent things with which they have no resemblance, no ‘rapport’, no intelligible relation. The historical situation thus seems to be as follows. The twin principles of causal and epistemological intelligibility (CIP and EIP) were inherited by Cartesian metaphysicians from their scholastic predecessors. Unfortunately, as we noted in Chapter Two, Descartes’ own metaphysics and epistemology seem to violate them. Malebranche, on my interpretation, is taking the ‘high priori road’, insisting that we retain CIP and EIP (though he doesn’t require resemblance in any naïve sense). Given CIP and EIP, he argues, only supernatural solutions will suffice: only God’s action can provide genuine necessary connections; only God’s ideas can provide genuine objective knowledge. Arnauld, by contrast, has less confidence in metaphysics and more respect for common sense. On his view we cannot doubt that our perceptions (the modes of our souls) are representations of objects; but equally, we cannot explain how they are, except by saying that God made them so. The crucial difference, in their metaphysics as in their theology, concerns the degree to which God’s acts and volitions must be intelligible to us. Why is Malebranche so confident that the modes of the human soul cannot be representations? He answers that every such mode is a 81
THE DISPUTE WITH ARNAULD
particular, and that no mere particular can represent something universal. And the idea of extension, étendue intelligible, is ‘infinite’ (in at least two distinct senses26), and therefore cannot be a mode of the finite human soul (OCM XVII–1 283). The obvious reply is made by Régis, who took up the defence of mainstream Cartesianism against Malebranche (OCM XVII–1 285). Intelligible extension, says Régis, is ‘objectively’ infinite but may still be ‘formally’ (i.e. intrinsically) finite, and hence can still be a mode of a finite soul. To block Malebranche’s argument, it seems, Régis need only re-emphasise the absolute distinction between formal (intrinsic) and objective (representational) reality, and deny the validity of arguments that illegitimately slide from one domain to the other. So does Malebranche’s argument depend on nothing more than a simple fallacy of self-predication? Is he simply assuming that the idea of F must itself be F? If so, there is little more to be said on the subject, and no defence to offer. But we have already seen Malebranche vigorously reject any such interpretation of étendue intelligible. Although it is the idea of length, breadth and depth, intelligible extension is emphatically not itself extended in three dimensions (OCM VI 242). But if it is not extended at all, it cannot be literally infinite in extent. When Malebranche says that étendue intelligible is infinite and the human soul is only finite, he cannot mean that the former is literally too big to fit into the latter. He defends only a principle of epistemological intelligibility EIP, not a principle of epistemological resemblance ERP. The idea must make the object known, and this need not involve resemblance in any crude sense.27 A mathematical equation may represent a circle as well as a diagram. So when Malebranche says that étendue intelligible is ‘infinite’ what he means – or at any rate, what he should mean – is that it is objectively infinite, i.e. infinite in its content. Our familiarity with intelligible extension makes possible a sort of unlimited power or capacity to construct, identify, and comprehend all the infinite varieties of all the infinite kinds of geometrical figures. His point is not that intelligible extension is literally too big but that its content is too rich to fit into our limited minds.28 The point is repeated in the first of the Méditations Chrétiennes. Your thoughts, explains the Word, can fly through the infinity of space: What, do you think you are big enough to contain within yourself the immense spaces you perceive? Do you think that your being can receive modifications which represent to you the actual infinite? Do you think that you even have the capacity to contain in yourself the idea of everything you can conceive in what is called an atom; for you conceive clearly that the 82
THE DISPUTE WITH ARNAULD
smallest part of matter that you imagine, being divisible ad infinitum, potentially contains an infinity of distinct figures and relations. (OCM X 16) The first two sentences of this passage suggest the crude thought that intelligible extension is simply too big for our souls. But the reference to ‘what is called an atom’ quickly supplies the necessary correction, and thus wards off misunderstanding.29 Malebranche’s point is that in thinking of any portion of matter, however small, we can think of it as consisting of an infinity of parts, and as entering into an infinity of relations with other bodies. This capacity, he argues, is only possible in virtue of the Vision in God.
What is at stake? Arnauld’s version of the debate When reading Arnauld’s VFI and Malebranche’s Réponse, the attentive reader will be struck by a puzzling and somewhat disquieting thought. The protagonists in this prolonged and heated controversy appear to disagree not just about the subject under debate (the nature of ideas) but also about exactly what philosophical issues are at stake and which of these issues are of central importance. This thought must have struck Steven Nadler between the publication of Arnauld and the Cartesian Philosophy of Ideas (1989) and that of Malebranche and Ideas (1992). In that short period of three years, Nadler changed his mind not only about his interpretation of Malebranche, as we saw in our previous chapter, but also about the primary focus of his controversy with Arnauld.30 The earlier book views the controversy more or less through Arnauld’s eyes; the later book adopts Malebranche’s point of view. Let us begin with Arnauld’s version of the issues at stake. For Arnauld, the debate is, first and foremost, about the role of ideas in perception. Arnauld sees himself as the champion of an act theory of ideas and of a direct realist account of our perception of external objects. He doesn’t use precisely these terms, of course, but his intentions are sufficiently clear. Malebranche, he thinks, has an object theory of ideas and an indirect realist account of perception. On Malebranche’s theory, says Arnauld, all we are ever directly or immediately aware of in perception is an idea, which is an object with an independent reality of its own, distinct both from the mind of the perceiver and from the external object (if any) which is the indirect or mediate object of the perception. Malebranche insists, of course, that it is still true to say that we perceive
83
THE DISPUTE WITH ARNAULD
bodies, but Arnauld denies that he is entitled to say this. On Malebranche’s theory, we are told, the soul is: … condemned, by an irrevocable sentence of this philosophy of false ideas, never to see any body through itself, present or absent, close or distant. I would even dare to remove the words through itself, and say absolutely that it is condemned never to see any body, as we will see in what follows.31 God wills that we perceive tables and chairs, trees and houses. But God’s will is both efficacious and economical. To invent an intermediate realm of ideas between the mind and its objects would be at best a clumsy means of achieving God’s purpose and at worst a way of actually frustrating it.32 What I see when I look at the sun, Arnauld insists, is the material sun itself, not a mere intelligible sun in the mind of God.33 Malebranche will say, of course, that the intelligible sun (or at least a portion of intelligible extension) is the direct or immediate object of my perception, while the material sun is its indirect or mediate object. But his language, Arnauld objects, is not consistent. Sometimes he writes that we see the ideas (in God) of things; sometimes that we see the things themselves. On his own principles, Arnauld argues, he should entirely abandon the language of realism. The immediate object of all perceptions must be, for Malebranche, something intimately united to our souls. God is capable of such union; bodies are not. On Malebranche’s principles, then, God is Himself, strictly speaking, the immediate object of all our perceptions. The Vision in God collapses into a Vision of God.34 Malebranche thus sets out to explain how we can see creatures, and ends up with a theory that entails that we see only God and not creatures. Arnauld also depicts himself as the defender of Descartes against the dangerous innovations of Malebranche. Descartes, Arnauld insists, never thought of ideas as distinct from perceptions. His most explicit discussion can be found in the Second Replies, which lends some support to Arnauld’s interpretation. For Descartes, ideas are modes of the soul, mental acts: an idea is ‘our thought itself, insofar as it contains objectively what is formally in the object’.35 On such a view, there is no need whatsoever to postulate a distinct realm of representative entities. When I am seeing the sun what I am aware of is the sun itself, insofar as it is ‘objectively’ present to my mind. Malebranche would reply that this is not so much a theory of intentionality as an anti-theory, an implicit rejection of the demand for
84
THE DISPUTE WITH ARNAULD
an explanation. If all we can say is that it is the nature of mental states to have representational content, and that this is so because God willed it, nothing whatsoever has been explained. Arnauld responds explicitly to this objection. It is a mark of wisdom, he suggests, to know where to stop in our search for explanations. We say that the soul thinks because it is its nature to think, or because God gave it the faculty of thought. Malebranche objects that this is non-explanatory. But, Arnauld replies, there are many contexts in which this is ‘the best reply that can be given’.36 If we ask why iron is magnetic, it is idle to posit a magnetic faculty when a superior (mechanistic) account is in principle possible. Cartesian physics eliminates talk of faculties in this sort of context. But if we ask why matter is mobile, all we can say is that that is its nature, i.e. that God has made it that way. All our physical theories presuppose this power of mobility, so no physical theory can possibly account for it. It is in this latter sense, Arnauld explains, that I say that the mind has the faculty of thought. Any attempt to explain how thought is possible, i.e. how the modes of the soul can be representations of actual and possible objects, will be dismissed by Arnauld as (a) a presumptuous piece of hubris on our part, and (b) an error derived from mistaken and misapplied physical analogies. Malebranche is immodest in not knowing when explanation has to stop, and muddled in thinking that models borrowed from physics will cast any light on the mind. Hence his ‘walking soul’ argument and his confusion about the notion of presence.37 Arnauld rejects not just any naively literal sense of resemblance (Watson’s ELP), but also any principle of epistemological intelligibility (our EIP). There is nothing about the intrinsic features of the mode of my soul that represents F that particularly fits it to be the idea-of-F rather than the idea-of-G. We can make no valid inferences from objective reality (representational content) to formal reality (intrinsic features) or vice versa. Intentionality must simply be accepted as a primitive and irreducible feature of the mental.38
What is at stake? Malebranche’s version of the debate For Malebranche, the correct analysis of perception (direct versus indirect realism) is a mere distraction, far from the centre of his concerns. He continues to maintain that we do see bodies, adding only the qualification that we do not see them en eux-mêmes (OCM VI 101). But all he is seeking to do here is to fend off Arnauld’s objections, to show that his version of indirect or representative realism does not collapse into
85
THE DISPUTE WITH ARNAULD
idealism, and that the Vision in God is not a Vision of God. What is important to Malebranche in the controversy is not the theory of perception but the very possibility of objective thought.39 If ideas are just modifications of our souls, Malebranche maintains, there will be no escape from the Pyrrhonian scepticism he saw gathering strength all around him.40 On inspecting my idea of X, I can say confidently that my idea represents X as having properties FGH, so I can affirm that I perceive X as FGH. But if my idea is just my perception, this is merely a judgement about how things seem to me, not how they are in themselves. I can’t infer anything about how X will appear to other finite minds, or about whether X objectively is as I perceive it. Every mind will remain trapped within its own ideas, with no escape from subjectivity. Malebranche clearly intends his theory of the Vision in God to block two related sceptical doubts. There is the inference from the subjective to the objective: how do I know that what seems F to me is objectively F? And there is the inference from the subjective to the intersubjective: how do I know that what appears F to me will appear F to others? The sceptic will challenge both inferences, insisting that we should only judge how things appear to us, not how they are in themselves or how they appear to other observers (if other observers exist). The Vision in God answers the first question directly, but also supplies as a corollary an answer to the second. By providing a guarantee of the objectivity of thought it also ensures intersubjectivity. If X is objectively F, any observer who perceives X clearly and distinctly will perceive it as F. Without ideas in my sense, Malebranche maintains, there is no escape from scepticism. He even attempts to enlist Arnauld’s own testimony by citing the famous Port Royal Logic. In the Art de Penser, the following principle is established as necessary to all certainty in the sciences: whatever is contained in the clear and distinct idea of X can be truly affirmed of X (OCM IX 924). If, for example, our idea of matter (three dimensional extension) represents it as infinitely divisible, then matter is infinitely divisible.41 But unless our ideas were the divine archetypes according to which things were (and are) created, there would be no reason to suppose that this principle was true. Things must conform to their divine archetypes, so the Vision in God provides knowledge of objectively necessary truths.42 If Arnauld is right and my ideas are mere modes of my soul, the principle would not hold, and we would have no escape from scepticism: A creature necessarily conforms to the idea of whoever created it. I agree. But the idea that you have of it, you say, is nothing 86
THE DISPUTE WITH ARNAULD
but a modification of your soul. And this modification is certainly not the Creator’s idea on the basis of which He formed this creature. It is thus in no way certain that the thing conforms to your idea, but only that you think it does. Thus, your view establishes Pyrrhonism, but mine destroys it. (OCM IX 925) If clear and distinct ideas compel assent, as Descartes has taught, then I may find myself unable to doubt that things are as my ideas represent them, but subjective certainty is no substitute for objective truth.43 The same point is repeated in the supplementary material appended to the Entretien d’un Philosophe Chrétien et d’un Philosophe Chinois (1708). To maintain that ideas are nothing distinct from perceptions, Malebranche claims, is to eliminate objective standards, and thus to establish Pyrrhonism in the sciences and libertinage in morals: This principle is generally accepted: That one can affirm of a thing what one clearly perceives to be contained in the idea one has of it. But if the idea which one has is not distinct from the perception, or from the modification of the mind which perceives it, this principle is not certain. For assuredly God has not created things on the basis of our fleeting perceptions, but on the basis of His eternal ideas: which ideas we perceive when they touch us, and by means of them we perceive the things which are necessarily in conformity to them. (OCM XV 50–1) The first principle of the sciences, according to the Cartesians themselves, is that we may affirm of the thing whatever is contained in the clear and distinct idea of it. But that principle would not be certain unless the Vision in God were true. So the Cartesians face a stark choice: abandon all claims to certainty in the sciences, or embrace the Vision in God. By solving the problem of the objectivity of thought, Malebranche also solves the problem of intersubjectivity. There is a single creator, and a single set of ideas that serve Him as archetypes. In geometry, there is one model or archetype, intelligible extension, equally accessible in principle to all minds capable of sufficient attention. In morals, there is one common notion of Order, equally accessible – albeit perhaps with rather more difficulty because of the need to still our passions and silence our prejudices – to all minds. Since all humans participate in a common
87
THE DISPUTE WITH ARNAULD
reason, there is no ground for relativism, either in mathematics or in morality.44 We can learn the arithmetic of the Hindus and the astronomy of the Chinese, and they can learn ours. Exactly the same must hold for the rational principles of morality. For Arnauld, by contrast, intersubjectivity could only hold Deo volente, as a result of a sort of divinely pre-established harmony between the contents of the perceptions of individual souls. Malebranche’s position might best be set out as a sort of transcendental argument in the Kantian manner.45 It starts from the assumption that objective thought is possible, and then enquires into the preconditions for that possibility. Schematically, we might present the argument thus: 1 2
Objective thought is possible. Objective thought is only possible if the Vision in God is true. Therefore, the Vision in God is true.
This is clearly valid, so criticism must focus on the truth of the premises. To defend (1) Malebranche would simply point to the existing sciences of mathematics and (more controversially) morality. That 2 + 2 = 4, and that the three angles of a triangle add up to 180°, are necessarily and absolutely true, not true on some condition or for some minds rather than others. That God is superior to a man, and a man to his dog, are for Malebranche just as clear examples of necessary and absolute truths. No element of subjectivity enters into the content of these thoughts, nor into the considerations which establish their truth. Critics of Malebranche would be well advised to grant (1) and challenge (2). This is Jolley’s strategy. He praises Malebranche for his ‘resolute anti-psychologism’,46 seeing in him a precursor of the views of Frege. Malebranche might have enjoyed, for example, reading the following passage from Frege’s famous article On Sense and Meaning: If every thought requires a bearer, to the contents of whose consciousness it belongs, then it would be the thought of this bearer only and there would be no science common to many, on which many could work. But I, perhaps, have my own science, namely a whole of thought whose bearer I am and another person has his … No contradiction between the two sciences would then be possible and it would really be idle to dispute about truth, as idle, indeed almost ludicrous as it would be for two people to dispute whether a hundred mark note
88
THE DISPUTE WITH ARNAULD
were genuine, where each meant the one he himself had in his pocket and understood the word ‘genuine’ in his own particular sense.47 But, Jolley continues, it is one thing to establish the existence of an objective ‘World Three’ of ideas (logical concepts) and propositions, and quite another to endorse the Vision in God.48 It is possible to endorse the first premise of Malebranche’s transcendental argument but to reject the second, and thus to escape his conclusion. We simply replace the mind of God with ‘World Three’, the realm of objective thought. There are of course well-known objections to such an extensive ontology of abstract objects, but are these difficulties any greater than those of Malebranche’s theism?
Deeper reasons Why, we asked at the beginning of this chapter, did Arnauld write des Vraies et Fausses Idées? His primary concerns were theological, and his most violent disagreement with Malebranche was over the subject of grace. If Malebranche had argued from his theory of ideas to his opinions on grace, it would be easy to understand why Arnauld should choose to undermine support for the latter by refuting the former, but there is no such explicit argument in Malebranche. He in fact accused Arnauld of simply changing the subject in choosing to attack the Vision in God rather than the account of grace in the Traité (OCM VI 18). The Vision in God, he explains, is the most abstract part of my work, and has no relation at all to the arguments of the Traité. Rather than deal with the arguments of the Traité on their merits, he protests, Arnauld has resorted to personal attack and unfair caricature, portraying me as a mere dreamer and a visionary. But this is disingenuous on Malebranche’s part. There is a connection between the theory of ideas on the one hand and the account of grace on the other. Although Malebranche insists that the issues are quite distinct, the more perceptive of the commentators disagree.49 On this point, says Alquié, we cannot follow Malebranche: The doctrine of ideas is indeed the foundation of the doctrine of grace, since the doctrine of grace assumes that we can know the principles of divine action. Now we can only do this through perceiving, by Reason, ideas in God.50
89
THE DISPUTE WITH ARNAULD
Denis Moreau helps to make the connection clearer in his admirable account of the controversy.51 The connection, he explains, is univocity: if our ideas are God’s ideas, then we can make confident judgements about God’s intentions and God’s reasons for acting. ‘We do see, in God and as God, our ideas; we are thus well-grounded in our evaluations of what God wants and does’.52 Since our ideas are God’s ideas, it is not, pace Arnauld, ‘rash’ or ‘presumptuous’ to claim some rational insight into God’s designs and His methods for realising those designs. In so doing, I am not boldly asserting my own human reason but humbly following universal reason, which is the reason of God Himself. As Malebranche puts it in the Traité, Were I not persuaded that all men are rational only because they are enlightened by the Eternal Wisdom, I would, without a doubt, be rash to speak about God’s plans and to want to discover some of His ways in the production of His work. But because it is certain that the Eternal Word is the universal Reason of minds, and that by the light that it shines on us incessantly we can all have some communication with God, I should not be blamed for consulting this Reason, which, although consubstantial with God Himself, does not fail to answer those who know how to interrogate it with serious attention. (OCM V 24–5) All knowledge of truth, Malebranche explains in Book Five of the Recherche, is a sort of union with God. When the mind discovers some truth, or sees things as they are in themselves, it sees things in God’s ideas, not in its own modes. (God is, but the soul is not, a monde intelligible.) So whenever the mind knows the truth, it is to that extent united with God and can therefore be said to know and possess God: But not only might we say that the mind that knows the truth to a certain extent also knows God who contains the truth, we might even say that to a certain extent it knows things as God knows them. Indeed, this mind knows their true relations, which God also knows. The mind knows them in the perception of God’s perfections that represent them, and God, too, knows them in this way. (OCM II 168, LO 364) Malebranche thus shares in what Edward Craig has called ‘the insight
90
THE DISPUTE WITH ARNAULD
ideal’ that is such a characteristic feature of seventeenth century rationalism.53 Our knowledge, at its best, equals God’s knowledge in its clarity and certainty though not of course in its extent. Mathematics provides, as ever, the most convincing examples of this rational insight, but it was not mathematical knowledge that proved contentious. For Malebranche, our rational insight extends not just to mathematics but to morals: through our perception of Order we are able to gain some insight into God’s designs, and thus even to criticise His creation. As Ariste exclaims at the end of the ninth Entretien, Ah Theodore, how well all your principles are connected! By what you have just told me I understand further that it is in God and in an immutable nature that we see beauty, truth, justice, since we are not afraid to criticize His work, to note the defects in it, and even to conclude thereby that it is corrupt. Indeed, the immutable order, which we see partly, must be the law of God Himself, written in His substance in eternal and divine characters, since we are unafraid of judging His conduct by the knowledge we have of that law. (OCM XII 221, JS 169) Arnauld thinks that it is presumptuous for men to claim sufficient knowledge of God’s intentions to criticise His works. He never tires of repeating the words of Isaiah, ‘My thoughts are not your thoughts’, and emphasising the distance and difference between our knowledge and God’s knowledge. 54 But without knowledge of God’s intentions, Malebranche retorts, we can neither praise His wisdom in creation nor recognise the effects of the fall. We praise the wisdom of an artificer when we see how well his work serves his design; if we are entirely ignorant of the design, we can’t assess the skill of the workman. Likewise, if we can’t trust our own judgements of value, we can’t see that our world is the mere ruin of a perfect creation. Order tells us, for example, that souls are superior to bodies, and that it is therefore a disorder, in a perfectly objective sense, for a man to live for his belly. But this argument relies on a conception of Order as something real and objective, binding alike on God and man. The superior must be preferred to the inferior; the inferior exists for the sake of the superior. Without such an Order, Malebranche argues, there would be no proof of morality and religion: For how would one prove to a libertine that nature is disordered, if there is no necessary and immutable Order? He has only to
91
THE DISPUTE WITH ARNAULD
reply boldly that God has made minds for the sake of bodies, for eating and drinking and the peaceful enjoyment of sensible objects. How will we prove to him that God will recompense good works and punish crimes, and even that justice and injustice are not phantoms that are used to strike fear into the credulous? The libertine has only to reply proudly and brutally that God’s wisdom and reason are quite different from ours: that it seems just to us to recompense what are called good works, but that what appears just is not in the least so, or is not in the least so with regard to God, who is the absolute master of His creatures: finally, that His wisdom and justice, if one wants to attribute such qualities to Him, have nothing in common with our feeble thoughts. (OCM VI, 23) The rational theologian, Malebranche maintains, must seek to silence the unbelievers, ‘faire taire les impies’. The unbelievers point to all manner of evils in our world, and argue that it could not be the creation of a wise and benevolent God. Where is the wisdom, they will ask, in making an animal so deformed that it cannot survive, or helping crops to grow only to ravage them by hail? Worse still, where is the wisdom and benevolence in establishing unjust princes, letting whole nations perish, and causing heresies to flourish at the expense of true religion? (For the freethinker, of course, this latter argument would be ad hominem.) The theology of Arnauld, Malebranche argues, can provide no answer at all to such doubts. Arnauld’s God is omnipotent but inscrutable, exercising His power by means of particular volitions. So on Arnauld’s theory God does intend – for utterly mysterious reasons of His own – each monster, each hailstorm and each unjust prince. Such a theology will not, Malebranche argues, convince the doubters: It is a question of silencing the unbelievers, and placing reason on the side of religion. It is not enough to satisfy the believers, we must put pressure on the freethinkers, and force them to keep quiet: or rather, we must convince them that God’s conduct is worthy of His wisdom and His goodness, which is very easy on my principles. (OCM VII 592) Malebranche sees his philosophical theology as a weapon for use not only against the freethinkers but also against Christian heretics (Protestants,
92
THE DISPUTE WITH ARNAULD
Socinians) and adherents of other religions (Jews, Moslems, Confucians). In the Conversations Chrétiennes, for example, we find an explicit discussion of the Trinity and the hierarchy of persons, against the more austere monotheism of the Socinians and of Islam. Two powerful objections of a Chinese bonze are also discussed. Christianity, says the bonze, is not a universal religion, so belief in Christ cannot be necessary to salvation. As for the Christian theory of eternal damnation, it is rejected by the bonze as inconsistent with God’s goodness.55 The Conversations can thus be regarded, as Robinet shows, as an early contribution to the field of comparative religion.56 And towards the end of his career Malebranche re-entered this field with his Entretien d’un Philosophe Chrétien et d’un Philosophe Chinois (1708). Written against the supposed ‘atheism’ of the Chinese (in fact, the position Malebranche ascribes to them is remarkably close to Spinozism), it also takes issue with Jesuit missionaries who were – in the eyes of many Christians – making too great concessions to Chinese ‘idolatry’ (e.g. ancestor worship) in order to gain large numbers of spurious converts. When Malebranche says that reason should be placed on the side of religion, he does not mean merely that – as the Church demanded – philosophers should provide proofs of the existence of God and the (natural) immortality of the soul, and then leave the rest to the theologians. He thinks that reason can go much further than this into the Christian mysteries. In the Conversations Chrétiennes, for example, he argues that only the incarnation of the Word makes the creation worthy of God (OCM IV 132–3). It follows that all religions other than Christianity can offer no credible account of creation. God is, after all, sufficient unto Himself: it is not as if He needed anything from the world He has created. But Malebranche’s attempts to extend philosophical reasoning into theology can easily backfire. Without the incarnation of the Word, says Malebranche, the world would not be worthy of God. Why not? Because there can be no relation or ‘rapport’ between the finite and the infinite. But if this is so, the freethinker will retort, there can be no Incarnation and no single person, Jesus Christ, with two natures, divine and human. Malebranche’s supposed solution tacitly helps itself to an assumption which is – on his own principles – strictly impossible. Similar difficulties will surface in the moral domain. If we trust our intuitions of Order, we will surely find ourselves endorsing the argument of the Chinese bonze – a just and benevolent God wouldn’t condemn the vast majority of mankind to eternal torment for the crime of a remote ancestor. Only a cruel and vindictive tyrant would do that. And if belief in Christ is necessary to salvation, what is to become of all those millions 93
THE DISPUTE WITH ARNAULD
who are excluded from the Christian Church by accidents of history and/or geography? Our intuitions of justice tell us clearly that it would be unfair to penalise such people, and justice is not one thing for God and another for man. Such arguments would become a prominent feature of Enlightenment deism. The controversy between Malebranche and Arnauld thus marks a fissure within Christian theology, a fundamental parting of the ways. Malebranche demands a theology that appeals to uniform and universal norms or standards, and seeks to defend orthodox Christian dogmas in terms of those universal norms.57 His arguments are designed to be culture-neutral, to appeal to all men whatsoever, simply in their capacity of rational beings. He can thus take on the freethinkers on their own terms, and persuade Moslem imams and Chinese bonzes of the superiority of Christianity – if the arguments work. But equally he leaves himself wide open to the counter-arguments of the freethinkers and of the spokesmen of other faiths. Arnauld, by contrast, wants only the most modest of philosophical theologies, confined perhaps to the Cartesian proofs of God and the soul. For further insights into matters of theology, we must turn to Scripture and to the authority of the Fathers. The division of philosophy from theology remains fixed and impassable. To claim that we can see the universe through God’s eyes is dangerous and presumptuous folly. For Malebranche, the introduction of philosophical rationalism into theology is not presumptuous or sacrilegious. Universal reason is itself the voice of the Word, so when we seek to ‘rationalise’ scripture, i.e. to make its message consistent with our metaphysical and moral intuitions, we are not taking it upon ourselves to correct God’s revealed word. Rather, we know for certain that God cannot contradict Himself, so His word as revealed in scripture must be consistent with His word as manifest through reason. Apparent contradictions must be explained away, e.g. some passages in scripture may be mere concessions to the level of understanding of the vulgar.58 Christian theology can in principle be taught from rational first principles, without reliance on the established opinions of the Fathers. This was Malebranche’s strategy in the Traité, and was one of the features of the work that gave most offence to critics such as Bossuet and Fénélon as well as Arnauld. Malebranche denies, of course, that he is rejecting the authority of the Fathers. Truth, he will say, is truth, whatever its source, and a short work of Christian theology resting only on rational first principles is far more likely to persuade the infidels than yet another weighty tome full of interminable quotations in archaic Greek and Latin (OCM V 3–6). 94
THE DISPUTE WITH ARNAULD
So when Arnauld chose to launch his long campaign against Malebranche with des Vraies et Fausses Idées, he knew exactly what he was doing. He saw the Vision in God – correctly – as the foundation of Malebranche’s philosophical theology. He also saw – equally correctly – that Malebranche’s rationalism represented a potential threat to many aspects of Christian tradition. Malebranche was of course confident that reason could be co-opted onto the side of religion in the battle against the unbelievers, and that the few small concessions that might prove necessary in the interpretation of scripture would not threaten anything fundamental to Catholic orthodoxy. But such confidence was naïve. By the time of Malebranche’s death in 1715, Pierre Bayle had already raised serious doubts about the harmony of reason and religion, and the deists were challenging – in the name of universal reason – orthodox Christian views of the Trinity, Providence, Miracles and the Afterlife. From Malebranche to Voltaire (via Bayle) is not as big a leap as it might appear.
95
O C C A S I O N A L I S M A N D C O N T I N U O U S C R E AT I O N
5 OCCASIONALISM AND C O N T I N U O U S C R E AT I O N
How not to think of occasionalism Malebranche’s occasionalism is sometimes represented – often on the basis of a superficial reading of a remark in Leibniz’s New System1 – as if it were merely an ad hoc solution to Descartes’ mind–body problem, with God intervening to fill the causal gaps between physical and mental events.2 This is a gross misrepresentation, based on a failure to understand both the full generality of Malebranche’s doctrine and the grounds on which it rests.3 It is not that he, like other Cartesian dualists, faces a problem in explaining how body and mind can interact, and is obliged to drag God in to plug the explanatory gaps. Rather, he has a perfectly general account of the nature of the causal relation which entails that only God can be a true cause, and this account yields, as a straightforward corollary, his solution to the mind–body problem. Occasionalism states that all so-called ‘second’ or ‘natural’ causes are not true causes at all, but serve merely as occasions on which the true cause (God) operates. As we saw in Chapter Two, earlier Cartesians such as Cordemoy and La Forge had articulated semi-occasionalist positions, usually denying causal powers to bodies. It is only in Malebranche, however, that we find a full-blooded occasionalism, denying all causal powers also to finite spirits.4 Only God, for Malebranche, has the power to bring anything about. But God does not act in random or arbitrary ways: on the contrary, He exerts His power in accordance with general rules, by means of what Malebranche calls volontés générales. So the impact of the white billiard ball does not, strictly speaking, cause the red ball to move (the white ball has no such power); the moment of impact merely serves as the occasion for God to redistribute motions according to certain universal laws. Similar stories can be told for all other so-called natural causes. The occurrence of certain events in my nerve organs and brain do not cause a sensation, but whenever my
96
O C C A S I O N A L I S M A N D C O N T I N U O U S C R E AT I O N
brain is in state φ God produces state Ψ in my mind. Likewise, my volition to raise my arm does not cause the motor nerves to fire, but serves as the occasion for God to produce that brain-state. Finally, my mind doesn’t even have any power over its own thoughts: attention is, for Malebranche, a species of prayer. I pray for illumination and, so long as my prayer is sincere, and persistent, and directed towards an appropriate object, God will reveal the required idea. The general thesis of occasionalism is stated clearly in the Recherche, where Malebranche seeks to show that there is only one true cause, and that all so-called ‘second’ or ‘natural’ causes are mere occasions. Because God acts in reliable, predictable ways, in accordance with universal rules, we will come to notice that events of type C regularly precede events of type E, will call the former ‘causes’ and the latter ‘effects’, and will come to expect an E-type event when we observe a C-type event. This is, of course, essential for everyday life and for the natural sciences, but has fatal consequences for metaphysics. (See OCM I 171–5, LO 76–8; OCM II 76– 84, LO 307–11.) The belief in natural causes is ‘the most dangerous error of the philosophy of the ancients’ (OCM II 309–20, LO 446–52), and the deep reason why the philosophy of Aristotle is not just un-Christian but anti-Christian. The Aristotelian acceptance of natural powers is, Malebranche argues, inseparable from pagan attitudes to nature. If one believes that the apple is the real cause of one’s pleasure, and that the sun is the sustaining cause of all vegetation, it is only reasonable to love the apple and to worship the sun. To refute this ‘miserable’ philosophy, the contrary truths must be established, namely that ‘there is only one true cause, because there is only one true God; that the nature or power of each thing is nothing but the will of God; that all natural causes are not true causes, but only occasional causes’ (OCM II 312, LO 448). The purpose of teaching occasionalism, in Malebranche’s eyes, is therefore moral and spiritual.5 Since humans naturally come to love whatever they regard as causes of their pleasures, Christians should be forever repeating the lessons of occasionalism, and seeking to learn them by heart. The Christian must love and fear only God, we are told in Éclaircissement XV of the Recherche (OCM III 211–52, LO 657–85), but this is impossible for a believer in second causes. The point is reiterated in the Conversations Chrétiennes (OCM IV 20). Whatever can cause us pleasure and pain, explains Malebranche’s spokesman Theodore, is a proper object of love and fear, so belief in natural causes leads inevitably to paganism and idolatry. Malebranche has two very different lines of argument in support of his occasionalism. There is what might be called the ‘thin’ argument, 97
O C C A S I O N A L I S M A N D C O N T I N U O U S C R E AT I O N
starting with the notion of cause as necessary connection, and arguing that only God’s will meets this condition. Then there is the ‘deep’ argument, taking its origin from the theological doctrine of continuous creation, which we touched on briefly in our discussion of Descartes in Chapter Two. The second and third sections of this chapter focus on the thin argument; the fourth and fifth sections set out the deep argument and its implications; the final section discusses objections and replies.
The idea of necessary connection Malebranche takes it for granted that a true cause is necessarily connected with its effect (OCM II 316, LO 448). What sort of necessary connection does he have in mind? Although he never spends much time on the topic, it is clear that he draws no sharp distinction between logical and metaphysical senses of necessity.6 The test for what is metaphysically possible is what can be conceived or described without self-contradiction. So for there to be a genuine causal relation between events C and E, one must be able to ‘see’ (i.e. rationally intuit) that C could not occur without E, i.e. that this would involve a contradiction. But this is never the case, Malebranche argues, for so-called ‘second’ or ‘natural’ causes. It is evident, he argues in Book Six of the Recherche, that no body has the force required to move itself, far less to set another body in motion (OCM II 312–13, LO 448). Our idea of body actually excludes such powers. So only a spirit can be a cause. But there is no necessary connection between the will of a finite spirit and its effects. We cannot discern any necessary connection between my willing to raise my arm and my arm going up. Indeed, we can see clearly that there is no such necessary connection: it is perfectly conceivable that the volition can occur without the action. So it seems that no finite spirit can move a body, any more than another body can. The search for necessary connections leads naturally into an analysis of our idea of power. (For Malebranche, as later for Hume, the ideas of power and of necessary connection are two sides of the same coin.) Previous generations of philosophers, says Malebranche, have spent enormous amounts of time and effort arguing about the powers of natural things, trying to define such terms as ‘force’, ‘power’, ‘efficacy’ and the like. But all such talk is, in the final analysis, unintelligible (OCM I 458–9, LO 242). The very notion of such a ‘force’, ‘power’, or ‘efficacy’, says Malebranche in Éclaircissement XV of the Recherche, is completely inconceivable. ‘Whatever effort I make in order to understand it, I cannot find in me any idea representing to me what might be the force or the power they attribute to creatures’ (OCM III 294, LO 658). The philosophers who believe in second
98
O C C A S I O N A L I S M A N D C O N T I N U O U S C R E AT I O N
causes are arguing without a clear idea of what they are talking about, which explains their endless controversies and confusions.7 Naturalistic philosophers, Malebranche explains, have been misled and seduced by their senses (OCM III 207–8, LO 659–60, and OCM III 232–3, LO 673–4). When they see plants growing after the rains, they assume that the rains have caused the vegetation, although they have no comprehension of how this could be. Averroes, for example, accuses the occasionalists of his day of having cracked brains for denying that fire burns and the sun illuminates, but his pretended ‘demonstration’ of these supposed causal powers amounts to nothing more than a blatant petitio principi. In fact, of course, occasionalists do not deny the manifest evidence of their senses. We need, Malebranche explains, to distinguish carefully between what is actually presented to the senses, and what is added by the mind of the observer, and may be the result of mere habit or prejudice, albeit a prejudice that seems to come naturally to us: When I see one ball strike another, my eyes tell me, or seem to tell me, that the one is truly the cause of the motion it impresses on the other, for the true cause that moves bodies does not appear to my eyes. But when I consult my reason I clearly see that since bodies cannot move themselves, and since their motor force is but the will of God that conserves them successively in different places, they cannot communicate a power they do not have and could not communicate even if it were in their possession. For the mind will never conceive that one body, a purely passive substance, can in any way whatsoever transmit to another body the power transporting it. (OCM III 208, LO 660) Strictly speaking, my eyes only seem to teach me that the first ball moves the second, or that sun and rain are the causes of vegetation. We need to consult reason, not to correct sense itself, but to correct an all-toohuman prejudice, a naïve tendency to confuse constant conjunction with genuine causation. In the Méditations Chrétiennes, this Malebranchian doctrine is voiced by no less an authority than the Word of God: Renounce your prejudices, my son, and never judge, with regard to natural effects, that one thing is the effect of another, because experience teaches you that it never fails to follow it. For of all false principles this is the most dangerous and the most fertile source of errors. As the action of God is always uniform and
99
O C C A S I O N A L I S M A N D C O N T I N U O U S C R E AT I O N
constant, because His volitions are immutable and His laws inviolable, if you follow this false principle, although God does all, you will conclude from it that He does nothing. (OCM X 59) As Pierre Bayle puts it, human beings are ‘drawn by their nature, and in some way instructed by a natural lesson, to judge that anything which is regularly connected with a certain effect, and without which that effect is not produced, is the true cause of it’ (OCM XVII–1 591). This natural tendency, harmless and even benign both in practical everyday contexts and in the natural sciences, is the source of dangerous errors in metaphysics. The natural judgements in question must be automatic, powerful, but resistible – if not, our Creator would be to blame for our errors. To correct the error, one must carefully distinguish what is actually ‘given’ to the senses from what is automatically added in judgement, and then subject this extra element of judgement to rational scrutiny. Reason, once consulted, corrects the error, but does not eliminate its source – we remain liable to go on making the same errors again and again. Occasionalism is a lesson we will forever have to learn anew and repeat to ourselves, day after day, without any hope of it becoming second nature. If a true cause must be necessarily connected with its effect, it is easy to show that only the will of God can be such a cause. There is a necessary connection between ‘God wills that X’ and ‘X occurs’, so long as God is conceived as omnipotent and X as logically possible (not selfcontradictory). This argument appears in Book Six of the Recherche (OCM II 313, LO 448) and is repeated in Part 1 of the Traité (OCM V 27). Here Malebranche draws the further inference that, since God’s will is sufficient for the willed effect to occur, He has no need of any instruments to bring about His designs – His volitions are, in and of themselves, efficacious. There would be a contradiction, explains the Word in the Méditations Chrétiennes, in God’s willing a man’s arm to move, and the arm not moving (OCM X 64). There is a necessary connection between the will of an omnipotent being and its effects. To deny this, explains the Christian philosopher in the Entretien d’un Philosophe Chrétien et d’un Philosophe Chinois, is simply to deny the existence of an omnipotent being (OCM XV 33). Critics of Malebranche, from Fontenelle8 and Hume9 down to Church10 and Watson,11 have objected to the tautological and seemingly merely verbal character of this argument. I understand, writes Fontenelle in his Doutes, that 100
O C C A S I O N A L I S M A N D C O N T I N U O U S C R E AT I O N
… since He [God] is all-powerful by His essence, it is impossible that He should will that a thing exist, and that that thing not exist. But do I conceive how that thing is, as soon as God wills that it exist? Not at all; on the contrary …12 I must grant that the will of God is a real cause because I can see that there is a necessary connection between it and its effects, although I don’t understand how it produces those effects. But can I not say exactly the same, asks Fontenelle, regarding the collision of bodies? I see that the red ball must move when struck by the white ball (because both are impenetrable) even though I cannot conceive how ‘force’ (whatever that may be) can be transferred from one body to another. Hume expresses this objection with characteristic clarity and vigour. If, he writes, we say that: … the idea of an infinitely perfect being is connected with that of every effect, which he wills, we really do no more than assert, that a being, whose volition is connected with every effect, is connected with every effect; which is an identical proposition, and gives us no insight into the nature of this power or connexion.13 We seek rational insight into the relation between cause and effect, and are fobbed off with what looks like a merely verbal proof. If I can’t understand how my volitions can bring about the voluntary motions of my body, does it help to be told that the volitions of an omnipotent being are, necessarily, efficacious? If I understand ‘omnipotent’, I will grant the validity of the inference from ‘An omnipotent being willed X to occur’ to ‘X occurred’, but it seems to shed no light whatsoever on the nature of the causal relation. Malebranche has a twin-track response to this fundamental objection.14 His main line of response is his theory of continuous creation, which we shall discuss later. He also seeks, however, to meet the objection more directly. In the ninth of the Méditations Chrétiennes he admits frankly that we have no clear idea of power, and thus cannot conceive how God creates the world, but he denies that this invalidates his argument (OCM, X, 97). The point is repeated in his reply to Fontenelle. To judge that A is the cause of B, he insists, we need only to see that A could not occur without B. One need not also see why this necessary connection holds, i.e. how A brings B about (OCM XVII–1 580).
101
O C C A S I O N A L I S M A N D C O N T I N U O U S C R E AT I O N
Particular causal relations We must now examine how Malebranche applies the general doctrine of occasionalism to particular types of causal relation: body to body (impulse), body to mind (sensation), mind to body (voluntary motion), and mind to mind (thought). Let us deal with each in turn, starting with the apparent power of a moving body to set another body in motion. We have already seen Malebranche’s denial, in Éclaircissement XV of the Recherche, that the mind can conceive of any genuine moving force or power, transferred from one body to another in collisions. In the fifth of the Méditations Chrétiennes, the author admits his embarrassment regarding his incapacity to form any clear idea of this supposed power. The Word promptly replies that, if a man will but still his senses and set aside his prejudices, the truth will become apparent to him. Suppose the physical universe consisted of only one cubic foot of solid matter: Could this body move itself? In your idea of matter, do you discover any power? You do not reply. But supposing that this body truly had the power of moving, in which direction will it go? How fast will it move? You are still silent? (OCM X 47) These rhetorical questions run together two clearly distinct arguments. First of all, there is an appeal to our clear idea of body or material substance. This idea, as Descartes taught us, contains only size and shape and the capacity for local motion – geometric and kinematic properties but no dynamic ones. But a clear idea, for Malebranche, excludes whatever it does not include. If I have a clear idea of X, I can tell a priori, for any property F, whether X is F or non-F. If the idea of X contains F, X is F; if it does not contain F, X is non-F. That is what it is, for Malebranche, to have a clear idea. This principle licences the inference from ‘My clear idea of body doesn’t represent it as possessing any dynamical properties’ to ‘My clear idea of body represents it as devoid of dynamical properties’. The second strand of argument involves an application of the principle of sufficient reason. If our hypothetical body had a motive power, and was thus capable of self-motion, the power would have to produce some determinate motion, in terms of speed and direction. But no sufficient reason could be provided in such a case. What happens, the Word continues, when a moving body strikes another body?
102
O C C A S I O N A L I S M A N D C O N T I N U O U S C R E AT I O N
… what will become of it, when it encounters another of which it knows neither the solidity nor the size? It will give to it, you will say, a part of its motive force? But who has taught you this? Who has told you that the other will receive it? What portion of this force will it give to it? And how will it be able to communicate or spread this force? Do you conceive all this clearly? (OCM X 48) Your senses, the Word continues, tell you only that the resting body begins to move after being struck by the moving body; to infer